Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Yes, it costs money, but it is way better than anything else due to its search capabilities. You get what you pay for. Said that, it is not really expensive. DEVONThink Personal costs US$ 50 and the Pro version costs US$ 80. They also have academic discounts.

Well not discounting that - but Mendeley itself has good searching to - you can search files names and also complete text internal to files.
 
Wow, your post is just huge. It appears that you are seriously worried about cross-compatibility. When you talk tri-platform, what do you mean? Mac, Windows and Linux?
Yes it is, I spent a long while typing it as I tried to recall what my exact software setup was since I reinstalled OS X :) Now it should have some structure for easier reading.

Overview of my setup
Mac OS X was and still is my primary platform, but I learnt with Lion that one should never take a workflow for granted, so at least full Mac/Linux compatibility of the core software was required. When I got this new Mac in the middle of the project, I expected it to work flawlessly as it did in SL. It didn't. At first I asked the Genius Bar to solve it. They refused, arguing they don't support non-Apple software. Then I read on the internet about "reverse-upgrading" to Snow Leopard; since it was fully compatible, I went down this road with the assistance of a dear friend's "BlackBook" (the friend, not the BlackBook).

I tried to go that way once, but I found it to be too complicated. In fact, I figured out that I don't need cross-compatibility. I don't use Linux and I plan to never use it. It may be free and open source, but it is not half as popular as it should be and it is used by 1% or 2% of the computers in the world, according to data tracking websites. So, I can live without it.
Sure it's complicated, really, as dual-platform compatibility usually entails Windows + Mac, but really I could have gotten away with Mac/Linux compatibility only, just like the way it was done at the major institute I, for a time, considered to integrate. It was an all-UNIX setup with a large Solaris calculations server fed by Macs connected to the data acquisition equipment. It was only natural researchers & developers used Macs as well.

On popularity
Popularity isn't a sign of performance or cost-effectiveness. Macs are less popular than PC by absolute numbers, but people will tell you here they still use it for various benefits simply not available on common PC. For statistics, Gnumeric was found out to be more precise than Excel on statistics, and I assumed Calc wouldn't have been better because it tries to emulate so much of Excel. The only stable version was on Linux, and sometimes I had the need to open other documents in the Ubuntu VM. It had to be compatible.

Ditto for OpenOffice and LibreOffice. It is cross-platform, works in Windows, Linux and Mac. But there is not a version for the iPad available, nor for Android tablets, nor a cloud version that I am aware of.
That is true. However I didn't have the iPad back then. However OpenOffice actually had a flavor available for iOS: NeoOffice allowed documents to be edited on i-devices, and was probably the most advanced open source office suite available for the Mac. MS Office haven't made its way to the iPad back then, there was no significance of a Microsoft-owned Cloud.

Yes, it may be open source, but that alone is not guarantee of anything. The ODT files are not a de facto standard as DOC and DOCX files came to be. Word is what everybody else uses, and if you save your files in DOCX you will never run the risk of losing the content. If you are really worried about cross-platform compatibility, have you tried to use a cloud solution such as Google Docs and Spreadsheets? Or even Microsoft Office, which has an online version that you can use from anywhere?
As far as I remember, Google Docs weren't much developed. Even now, they're slow, don't allow for loading the university's thesis model, don't allow for building giant spreadsheets. And they're slow. Plus, were they been present, I categorically (some say dogmatically) refuse to use Google's services to store mission-critical data providing them truckloads of personal information about me.

On so-called "standards"
De facto standards is a perversion of the term "standards". It's not because a majority use it that it has to be recognized as standard. While I was writing, Word versions didn't default to DOCX format for saving files, as 2007 wasn't present on all machines. DOC was a compromise I wasn't willing to accept: transfer, edition, re-saving on different versions of Word always lead to inconsistencies in formatting, mysterious disappearance of paragraphs, reverse-highlighting of editions when they were displayed (!).

Caught in a formats war
As I had no time nor will to invest any manually correcting problems that should have never happened, I followed a real standard in the ISO sense I knew would work reliably across different machines loaded with different office suites versions. I took care to use only fonts easily available on all OSes, for example. It happened that MS Office, in typical Microsoft fashion and despite a rather accessible ODF format "whitebook", first couldn't save ODF at all, then required a plugin, and finally got it as a native capability, but always screwed up ODT formatting. MS also famously counter-attacked in the ISO standardization battle by publishing the OOXML specification, one ten times as large as ODF's, which was claimed to be so bloated only to prevent competitors from implementing it efficiently. The ISO obviously failed on this one, since the point of a standard is precisely to make it unnecessary and counter productive to have more than one for a given result. Even Microsoft hadn't fully implemented its own Office Open XML standard: it's planned for the next release of MS Office.

Microsoft takes pride into having its own formats, incompatible from one version to another, claiming "must-have" improvements; fine, I take pride in following standards (up to the paper format, although only one printed version was required).

Since then, I kept the habit of sending two documents when requested for edition: one ODT for proofreading, one PDF to show what it should look like should anyone have complains or doubts.

It is possible to stick with Free Open Source Software
In the end it allows me to claim that, yes, it is possible to make graduate studies using only (F)OSS, and that yes, it is efficient enough, and yes, it is more flexible than other solutions. Of course a balance had to be struck, probably influenced by my experience with computers during this MSc. First, just after UPS stole my failed laptop PC (or they lost it, but refusing to pay insurance until I brought them to court), I asked my advisor for a laptop computer to use for that project, my own being a failing homebuilt machine running Ubuntu, as I felt it wouldn't hold for long.

Computers timeline
I got one old cranky Thinkpad (Centrino, 256MB RAM. Even at that time it was almost insufficient to run Windows XP properly with all the data load, and was too slow in Ubuntu) that worked well for about three months, then in three days, battery wouldn't charge, hard drive failed with a characteristic noise, machine then wouldn't boot, then wouldn't light at all when plugged in. The lab's IT guy quoted a rather high $1.5K for repairs my advisor refused to pay, despite having received a $500K grant the same year for the lab. That was a seasoned lab-used computer. After much anger and my refusal to buy a machine on my own unless my stipend was exceptionally increased by a similar amount (I had to pay living and tuition on $10K a year and forbidden to take a job, which kept me in the red for the duration of the project), I was assigned to a desktop machine, told it was good enough. As soon as I booted and made a quick hardware review, it had Win95 inside, 2-ish GiB HDD, and maybe 32MiB RAM. In short, totally unusable for any office & heavy analysis job. Weeks pass, then I finally land a spot at his colleague's lab on a modern machine. Windows XP I used to grow the spreadsheet I started and painfully edited on my outdated Celeron 500, transferred to the homebuilt one, now I could move at a decent pace, but soon enough I hit the limits of this OS. Had to ask for RAM increase to 4GiB, got it, but Windows couldn't use it efficiently, then had to install Ubuntu to keep on working. Two weeks downtime while scrambling to find a solution to get a working video card, finally got it stable for a while before my contract ended and had to find another spot. That's when I got the first MacBook and learnt never ever to trust any hardware. Worked fine for a while, but when summer came I discovered a bug the Genius Bar wasn't able to solve, after a long series of network-related problems. My first Mac was a lemon. Got a MBP as a replacement, used it for a long time, sold it at a decent price despite obvious damage as I needed free cash to pay bills before buying another Mac. The latter came with Lion, which blocked my fragile workflow.

Now you should understand why I was trying make documents as compatible as possible. Nothing was ever certain until the very end, and I could end up on a completely different machine. Even now, I think this bad experience took its toll as I maintain a second, much older Mac ready for use.

As I do have a Mac, I want to take advantage of it. I want to use Mac-only software if such software is superior to the cross-platform alternatives.
Sure it's a very valid concern of yours, but you're limited to whatever software other people you're communicating with are using. Having a single, compatible exit-point to maintain is the least of two evils, and now that all Word versions allow for saving -X documents, it shouldn't be a big deal for you not to have it. Plus, I read that there are less cross-compatibility issues between Word versions and platforms than years past, so I don't exclude giving it another try. Still, I am not comfortable throwing more money in MS's pockets for a software that can't even be properly installed, even following the manufacturer's instructions (I'm looking at you, Office 365, though the worst is probably SPSS).

Support for rare software
Another problem with powerful but seldom-used software is you don't get as much support when an issue arises. Having to exchange too many emails with any company to solve a software issue is something I precisely wanted to avoid, and feel would nullify the benefit of having a well-supported and standards-compliant Mac.

Cost of moving & proprietary applications/formats
There's also a cost of moving: should I choose to replace Zotero by Sente, Writer by Word, Calc & SciDAVis by Excel, PSPP & Gnumeric by R or SPSS, downtime would be significant from having to manually verify each and every document, not mentioning a significant learning time. Proprietary applications have no interest in helping you move to and from other providers, they want to keep you locked and hooked to their product. Whenever these apps cease to be updated, or you lose the license to use them (as may be the case with MS Office that started to be provided for free for grad students), documents are locked and can be considered lost. This happened with an internally-developped app that was necessary to decode data files. Only, file format was completely proprietary, saving files by back-writing to the original, but was also was unstable, crashed frequently, corrupting the file in the process. I was told no documentation was left when the original programmer left, that the file format I had to extract hasn't been used for more than 2 years, that the new engineer-of-all-trades could reverse-engineer it given about 3 weeks and money, and build a partially-open-sourced Matlab application to manage it, but at the same time told they wouldn't invest a dime in that project. Again, open-source would have saved many wasted weeks struggling with an unstable software, and perhaps recover data files deemed corrupted. In the strictest sense, interoperability is more important than pure open-source. Mendeley is such a software: it works on all common OSes, but falls short on office suite integration. Same for Antidote: works on all common OSes, but as I recently contacted them about broken Wikipedia page display. Was told, basically, shell out another $100 for the new version, or live with it.

How scientists circumvent the moving cost: they don't move.
That's why researchers stick to old habits of keeping everything intact, including 30-year-old computers if they ever have to read even one or two 8-inch floppies. The closest you get to the experimental level, the more newer technologies are frowned upon unless they allow for a whole new kind of experiments. That's why even decades-old scientific equipment is very carefully maintained, loaned to other scientists under strict rules, and so highly valued despite obvious wear. They don't tend to like updating to new supports, or formats, yet need to keep data accessible for review as an academic requirement; not that it's assiduously followed, but unanalyzed data on unreadable supports or in unknown formats will never be retrieved, and potential discoveries won't be made. Such archaisms would have been non-existent with open formats. They build slower but stabler workflows. Articles are not usually organized on computer. Most researchers I know except maybe the youngest ones print each and every article on paper, and rely on their visual memory to find it in meter-high piles. Even my own research isn't that well organized. Sure everything is in one folder, but I doubt anybody who may want to build upon my data would be able to do so easily, not from using proprietary formats, but from using a quick-and-dirty structure. And there are many, unreferenced articles I keep on paper.

At another lab, data was stored on specially-encoded VHS tapes (retrieved according to a thermal paper trace, boxes of it), acquired in a specifically-built standalone Red Hat for preliminary analysis (I'll always remember the "$50k software only 5 people in the world are using" sentence said by a cynic advisor who refused to modify a single setting in that system), the files sent to another RedHat for FTP serving to the final stage, a Windows XP machine he didn't want to install SP2 on for fear of it breaking, you guessed it, MS Office.

Of course these requirements may sound completely foreign to you, as you're not dealing with data acquisition but with a strong emphasis on paper management.
 
Last edited:
Yes it is, I spent a long while typing it as I tried to recall what my exact software setup was since I reinstalled OS X :) Now it should have some structure for easier reading.

Wow! These posts are becoming larger and larger. I guess academics like to write more than common people. :D

Overview of my setup
Mac OS X was and still is my primary platform, but I learnt with Lion that one should never take a workflow for granted, so at least full Mac/Linux compatibility of the core software was required. When I got this new Mac in the middle of the project, I expected it to work flawlessly as it did in SL. It didn't. At first I asked the Genius Bar to solve it. They refused, arguing they don't support non-Apple software. Then I read on the internet about "reverse-upgrading" to Snow Leopard; since it was fully compatible, I went down this road with the assistance of a dear friend's "BlackBook" (the friend, not the BlackBook).

Sure it's complicated, really, as dual-platform compatibility usually entails Windows + Mac, but really I could have gotten away with Mac/Linux compatibility only, just like the way it was done at the major institute I, for a time, considered to integrate. It was an all-UNIX setup with a large Solaris calculations server fed by Macs connected to the data acquisition equipment. It was only natural researchers & developers used Macs as well.

My dual-platform compatibility entails Windows and Mac.

I have been a PC guy (DOS/Windows) for 15 years. Not by choice, but because I live in Brazil, and Macs were extremely rare here for several years. In 2008, I bought my first Mac, which was a white MacBook. In 2010, I bought a Windows desktop and in 2011 I bought a Windows laptop. In 2013, I bought my current laptop, which is a 15-inch retina MacBook Pro, and it is the best computer I have had so far.

I do not discard having to go back to Windows someday. In fact, I would be able to do everything I need under Windows. I wrote my Master's thesis with Windows software, and also my PhD thesis. Windows is a perfectly capable environment for me. And it must be. Here in Brazil, prices of computers are just incredibly high due to the abusive taxes charged by a very hungry, inflated and corrupt government. And the government also plays a lot with the floating exchange rates, in order to benefit some specific sectors of the economy. As a result of all this, a 15-inch retina MacBook Pro which costs US$ 2,000 in the U.S. will cost about US$ 4,500 in Brazil. Due to these prices, I understand that it is not impossible that I am forced to go back to a low-end Windows laptop for some time. So, my plan B always included a Windows laptop and I must keep compatibility with it.

I have tried Linux, but I just gave it. Linux is not for me. I tried several flavors, but I found it to be too annoying.

On popularity
Popularity isn't a sign of performance or cost-effectiveness. Macs are less popular than PC by absolute numbers, but people will tell you here they still use it for various benefits simply not available on common PC. For statistics, Gnumeric was found out to be more precise than Excel on statistics, and I assumed Calc wouldn't have been better because it tries to emulate so much of Excel. The only stable version was on Linux, and sometimes I had the need to open other documents in the Ubuntu VM. It had to be compatible.

Macs are less popular than PCs for several reasons, not necessarily related to cost-effectives. However, the PC has certainly a better cost-effectiveness ratio in the enterprise.

That is true. However I didn't have the iPad back then. However OpenOffice actually had a flavor available for iOS: NeoOffice allowed documents to be edited on i-devices, and was probably the most advanced open source office suite available for the Mac. MS Office haven't made its way to the iPad back then, there was no significance of a Microsoft-owned Cloud.

I was not aware of NeoOffice for iPad. Is it not available anymore?

As far as I remember, Google Docs weren't much developed. Even now, they're slow, don't allow for loading the university's thesis model, don't allow for building giant spreadsheets. And they're slow. Plus, were they been present, I categorically (some say dogmatically) refuse to use Google's services to store mission-critical data providing them truckloads of personal information about me.

I don't use Google Docs either.

On so-called "standards"
De facto standards is a perversion of the term "standards". It's not because a majority use it that it has to be recognized as standard. While I was writing, Word versions didn't default to DOCX format for saving files, as 2007 wasn't present on all machines. DOC was a compromise I wasn't willing to accept: transfer, edition, re-saving on different versions of Word always lead to inconsistencies in formatting, mysterious disappearance of paragraphs, reverse-highlighting of editions when they were displayed (!).

It may be true. However, it probably does not matter for the average user. People do not want recognized standards. People want to send a file and have this file read and understood by other people. That is what standards should be about.

I had a big problem in the past with Microsoft Word 2003. I was writing my Master's thesis and, just two months before the deadline, Word crashed and suddenly my 250-page .DOC was corrupt. Word crashed every time I opened the file. I tried to save another file. I tried to copy the content to another Word file. It did not work. Word crashed again. And again. I tried everything. And Word kept crashing. Panic.

What saved me was OpenOffice.org (it was called so back then). I tried to open the file in OpenOffice.org and it worked without crashing! Then I used OpenOffice.org to write the rest of my thesis. But the experience was not the best: the interface could be much improved and I missed the functionality of Word. Still, OpenOffice.org saved my life.

I moved to Word 2007 when it launched and it was certainly better.

Caught in a formats war
As I had no time nor will to invest any manually correcting problems that should have never happened, I followed a real standard in the ISO sense I knew would work reliably across different machines loaded with different office suites versions. I took care to use only fonts easily available on all OSes, for example. It happened that MS Office, in typical Microsoft fashion and despite a rather accessible ODF format "whitebook", first couldn't save ODF at all, then required a plugin, and finally got it as a native capability, but always screwed up ODT formatting. MS also famously counter-attacked in the ISO standardization battle by publishing the OOXML specification, one ten times as large as ODF's, which was claimed to be so bloated only to prevent competitors from implementing it efficiently. The ISO obviously failed on this one, since the point of a standard is precisely to make it unnecessary and counter productive to have more than one for a given result. Even Microsoft hadn't fully implemented its own Office Open XML standard: it's planned for the next release of MS Office.

Microsoft takes pride into having its own formats, incompatible from one version to another, claiming "must-have" improvements; fine, I take pride in following standards (up to the paper format, although only one printed version was required).

Since then, I kept the habit of sending two documents when requested for edition: one ODT for proofreading, one PDF to show what it should look like should anyone have complains or doubts.

I use .DOCX (OOXML). I know Microsoft did not implement it entirely when launched Office 2007, and I guess it only did with Office 2013.

However, I must use OOXML. Every publisher in my area accepts DOCX, and most of them only accept DOCX. Some of them accept other file formats, but I don't want to limit the number of journals that will consider my paper. So, I cannot just save in other format such as ODF or whatever I want. I must use DOCX.

It is possible to stick with Free Open Source Software
In the end it allows me to claim that, yes, it is possible to make graduate studies using only (F)OSS, and that yes, it is efficient enough, and yes, it is more flexible than other solutions. Of course a balance had to be struck, probably influenced by my experience with computers during this MSc. First, just after UPS stole my failed laptop PC (or they lost it, but refusing to pay insurance until I brought them to court), I asked my advisor for a laptop computer to use for that project, my own being a failing homebuilt machine running Ubuntu, as I felt it wouldn't hold for long.

Yes, it is possible to stick with open source software, if you have to, or if you want to make this choice. I don't. I happily pay for software when I think it is worth it.

Computers timeline
I got one old cranky Thinkpad (Centrino, 256MB RAM. Even at that time it was almost insufficient to run Windows XP properly with all the data load, and was too slow in Ubuntu) that worked well for about three months, then in three days, battery wouldn't charge, hard drive failed with a characteristic noise, machine then wouldn't boot, then wouldn't light at all when plugged in. The lab's IT guy quoted a rather high $1.5K for repairs my advisor refused to pay, despite having received a $500K grant the same year for the lab. That was a seasoned lab-used computer. After much anger and my refusal to buy a machine on my own unless my stipend was exceptionally increased by a similar amount (I had to pay living and tuition on $10K a year and forbidden to take a job, which kept me in the red for the duration of the project), I was assigned to a desktop machine, told it was good enough. As soon as I booted and made a quick hardware review, it had Win95 inside, 2-ish GiB HDD, and maybe 32MiB RAM. In short, totally unusable for any office & heavy analysis job. Weeks pass, then I finally land a spot at his colleague's lab on a modern machine. Windows XP I used to grow the spreadsheet I started and painfully edited on my outdated Celeron 500, transferred to the homebuilt one, now I could move at a decent pace, but soon enough I hit the limits of this OS. Had to ask for RAM increase to 4GiB, got it, but Windows couldn't use it efficiently, then had to install Ubuntu to keep on working. Two weeks downtime while scrambling to find a solution to get a working video card, finally got it stable for a while before my contract ended and had to find another spot. That's when I got the first MacBook and learnt never ever to trust any hardware. Worked fine for a while, but when summer came I discovered a bug the Genius Bar wasn't able to solve, after a long series of network-related problems. My first Mac was a lemon. Got a MBP as a replacement, used it for a long time, sold it at a decent price despite obvious damage as I needed free cash to pay bills before buying another Mac. The latter came with Lion, which blocked my fragile workflow.

Wow. What a story.

Now you should understand why I was trying make documents as compatible as possible. Nothing was ever certain until the very end, and I could end up on a completely different machine. Even now, I think this bad experience took its toll as I maintain a second, much older Mac ready for use.

In this case, it makes sense to use software that saves in filetypes that can be opened in as many systems as possible. Still, it is always possible to use OOXML and keep compatible.

Sure it's a very valid concern of yours, but you're limited to whatever software other people you're communicating with are using. Having a single, compatible exit-point to maintain is the least of two evils, and now that all Word versions allow for saving -X documents, it shouldn't be a big deal for you not to have it. Plus, I read that there are less cross-compatibility issues between Word versions and platforms than years past, so I don't exclude giving it another try. Still, I am not comfortable throwing more money in MS's pockets for a software that can't even be properly installed, even following the manufacturer's instructions (I'm looking at you, Office 365, though the worst is probably SPSS).

I want to try and use software which was designed for the Mac so I can have the whole experience and not limit myself to cross-platform alternatives. The Mac is an expensive machine, and it wouldn't make sense to have it just to use the very same software I would be using on a PC.

However, I want to keep compatibility. I have Microsoft Office 365 and I must say Office for Mac is much worse than Office for Windows. I can deviate from Office, but I must use software that will allow me to save in a format that Microsoft Office can open. I will do my last revision of my papers in Word. Always.

Support for rare software
Another problem with powerful but seldom-used software is you don't get as much support when an issue arises. Having to exchange too many emails with any company to solve a software issue is something I precisely wanted to avoid, and feel would nullify the benefit of having a well-supported and standards-compliant Mac.

Very true.

Cost of moving & proprietary applications/formats
There's also a cost of moving: should I choose to replace Zotero by Sente, Writer by Word, Calc & SciDAVis by Excel, PSPP & Gnumeric by R or SPSS, downtime would be significant from having to manually verify each and every document, not mentioning a significant learning time. Proprietary applications have no interest in helping you move to and from other providers, they want to keep you locked and hooked to their product. Whenever these apps cease to be updated, or you lose the license to use them (as may be the case with MS Office that started to be provided for free for grad students), documents are locked and can be considered lost. This happened with an internally-developped app that was necessary to decode data files. Only, file format was completely proprietary, saving files by back-writing to the original, but was also was unstable, crashed frequently, corrupting the file in the process. I was told no documentation was left when the original programmer left, that the file format I had to extract hasn't been used for more than 2 years, that the new engineer-of-all-trades could reverse-engineer it given about 3 weeks and money, and build a partially-open-sourced Matlab application to manage it, but at the same time told they wouldn't invest a dime in that project. Again, open-source would have saved many wasted weeks struggling with an unstable software, and perhaps recover data files deemed corrupted. In the strictest sense, interoperability is more important than pure open-source. Mendeley is such a software: it works on all common OSes, but falls short on office suite integration. Same for Antidote: works on all common OSes, but as I recently contacted them about broken Wikipedia page display. Was told, basically, shell out another $100 for the new version, or live with it.

It may be true that proprietary software may want to keep you locked. If the software is gone, then you will probably have to re-make it all over again.

However, I must make two points here:

(1) If the proprietary software is wildly popular, then another software (either proprietary and/or open source) will probably provide some sort of functionality to import the files into the new application. This would allow the new software to get more users.

(2) You also run risks with open source software. I have seen several times the developer of open source software cease development. If someone in the community wants, he may keep the project alive, but will he? I have seen several open source software be lost in time because of this.

How scientists circumvent the moving cost: they don't move.
That's why researchers stick to old habits of keeping everything intact, including 30-year-old computers if they ever have to read even one or two 8-inch floppies. The closest you get to the experimental level, the more newer technologies are frowned upon unless they allow for a whole new kind of experiments. That's why even decades-old scientific equipment is very carefully maintained, loaned to other scientists under strict rules, and so highly valued despite obvious wear. They don't tend to like updating to new supports, or formats, yet need to keep data accessible for review as an academic requirement; not that it's assiduously followed, but unanalyzed data on unreadable supports or in unknown formats will never be retrieved, and potential discoveries won't be made. Such archaisms would have been non-existent with open formats. They build slower but stabler workflows. Articles are not usually organized on computer. Most researchers I know except maybe the youngest ones print each and every article on paper, and rely on their visual memory to find it in meter-high piles. Even my own research isn't that well organized. Sure everything is in one folder, but I doubt anybody who may want to build upon my data would be able to do so easily, not from using proprietary formats, but from using a quick-and-dirty structure. And there are many, unreferenced articles I keep on paper.

At another lab, data was stored on specially-encoded VHS tapes (retrieved according to a thermal paper trace, boxes of it), acquired in a specifically-built standalone Red Hat for preliminary analysis (I'll always remember the "$50k software only 5 people in the world are using" sentence said by a cynic advisor who refused to modify a single setting in that system), the files sent to another RedHat for FTP serving to the final stage, a Windows XP machine he didn't want to install SP2 on for fear of it breaking, you guessed it, MS Office.

Of course these requirements may sound completely foreign to you, as you're not dealing with data acquisition but with a strong emphasis on paper management.

Yes, it is all true.
 
Thanks for the post.

How exactly do you do that? ... By the way, do you use Sente for citations?

Yes, I use Sente for citations. You can use special characters in the citation tags to deviate from the standard citation format. I have written about this in my blog post on Sente (http://www.joachim-scholz.com/academipad/2012/02/06/annotate-pdfs-sente-mac/); here is an excerpt:

You can either jump back to Sente and add a reference thought the toolbar icon or via the Command+Y shortcut, but you can also just cite from within your word processor by putting you the reference you have in mind into a delimiters (e.g., [Scholz 2011]). I prefer using the latter method, as it doesn’t require me to interrupt my writing flow. Also, I changed the default setting of the delimeter from {} to [] because it is easier to type (however, I had to use {} in the example below because of html coding, oh well…).

Once you have finished your document, just add a [bibliography] tag to the end of your document and tell Sente to scan it. Like magic, Sente will replace all your reference shortcuts with the proper citations and add a reference list. Sente will be smart to follow citation guidelines (e.g., changing from “Scholz, Smith, and Mitchell 2013″ to “Scholz et al. 2013″ after the first citation), and you can select what citation style you want to apply from an impressively long list of built-in formats (you can also modify/create your own style in the Bibliography Formats editor).

But hey, what if I have special needs when citing papers, such as adding page numbers (“Scholz 2011, p. 397″) or when I only cite the year (“As mentioned by Scholz (2011)”)? No worries, Sente has you covered! You can add modifiers (e.g., Scholz 2011@397, %Scholz 2011) to your citations, and Sente will adapt the way in which it displays the citation. You can find out more about these modifiers in the Sente manual under “Modifying In-Text Citations” (p. 261).
 
Yes, I use Sente for citations. You can use special characters in the citation tags to deviate from the standard citation format. I have written about this in my blog post on Sente (http://www.joachim-scholz.com/academipad/2012/02/06/annotate-pdfs-sente-mac/); here is an excerpt:

You can either jump back to Sente and add a reference thought the toolbar icon or via the Command+Y shortcut, but you can also just cite from within your word processor by putting you the reference you have in mind into a delimiters (e.g., [Scholz 2011]). I prefer using the latter method, as it doesn’t require me to interrupt my writing flow. Also, I changed the default setting of the delimeter from {} to [] because it is easier to type (however, I had to use {} in the example below because of html coding, oh well…).

Once you have finished your document, just add a [bibliography] tag to the end of your document and tell Sente to scan it. Like magic, Sente will replace all your reference shortcuts with the proper citations and add a reference list. Sente will be smart to follow citation guidelines (e.g., changing from “Scholz, Smith, and Mitchell 2013″ to “Scholz et al. 2013″ after the first citation), and you can select what citation style you want to apply from an impressively long list of built-in formats (you can also modify/create your own style in the Bibliography Formats editor).

But hey, what if I have special needs when citing papers, such as adding page numbers (“Scholz 2011, p. 397″) or when I only cite the year (“As mentioned by Scholz (2011)”)? No worries, Sente has you covered! You can add modifiers (e.g., Scholz 2011@397, %Scholz 2011) to your citations, and Sente will adapt the way in which it displays the citation. You can find out more about these modifiers in the Sente manual under “Modifying In-Text Citations” (p. 261).

Thank you. This is very helpful. Sente seems like a very good citation manager indeed. I like its interface and apparently it handles citations well and easily too.
 
With regards to formats, LibreOffice can save and open files in .doc or .docx formats. I have never noticed any problems when opening them in Word. You may be aware of this, but I haven't seen it mentioned in the thread so far.
 
With regards to formats, LibreOffice can save and open files in .doc or .docx formats. I have never noticed any problems when opening them in Word. You may be aware of this, but I haven't seen it mentioned in the thread so far.

I know that LibreOffice can open and save both .doc and .docx. Actually, Apple Pages can open and save .doc and .docx as well.

But none of them are perfect.

There is a known issue with LibreOffice Writer and footnotes. Footnotes are handled differently in Writer and Word in a way that a blank space is added before the footnote begins. As far as I am concerned, this issue has never been addressed. It may not seem a big issue. But my PhD thesis had over 1,000 footnotes and I just can't manage to fix them one by one. So, LibreOffice is not an option for me due to this.

Apple Pages could be OK, but it lacks features. It does not have cross-references which is a must-have feature for me.
 
Wow! These posts are becoming larger and larger. I guess academics like to write more than common people. :D
Even in one's non-native language! I learnt to conserve my energies for worthwhile subjects, and this is one.

I have been a PC guy (DOS/Windows) for 15 years. Not by choice, but because I live in Brazil, and Macs were extremely rare here for several years. In 2008, I bought my first Mac, which was a white MacBook. In 2010, I bought a Windows desktop and in 2011 I bought a Windows laptop. In 2013, I bought my current laptop, which is a 15-inch retina MacBook Pro, and it is the best computer I have had so far.
I've also read that, for some strange fiscal reason, Apple products were heavily taxed on import, but not necessarily common PCs. Even recently iPod Touch has been almost twice as expensive in Brazil than the rest of the world. Can you share more information about this?

I used Windows daily from 1999 (first computer ever) to 2006, then Ubuntu from 2006 to 2008, then OS X from 2008, though not exclusively.

I do not discard having to go back to Windows someday.

Due to these prices, I understand that it is not impossible that I am forced to go back to a low-end Windows laptop for some time. So, my plan B always included a Windows laptop and I must keep compatibility with it.
Can't deny I would never have to use it again, too, especially if the difference with common PCs was so high here as in Brazil. But here, an entry-level Mac can be had for $1500 after taxes, while a decent PC costs $1000. The difference wasn't great enough to justify lost productivity on inferior software / hardware / service, especially on something as critical as a thesis. Not to say I don't have a PC, I do, an older LG with about the same power as a 2009 MBP (same parts inside). I just don't use it because it feels so old and slow, even compared to my white MB, not mentioning noisy. It sits in a drawer most of the time, plus I don't think it would support Windows 7 properly, hardware-wise. I can be mistaken, but coming from 8 years of dominant Unix-likes use, the transition to Windows would feel like a big, backward step. In fact, I call all my Windows installations "legacy". This pretty much sums it. And a PC can run Linux of course, so I don't feel tied by Windows. The only moment I still use it: on the job, when the Macs are taken; runs on a buggy machine, full of spyware, and dated hardware. Maybe that's why I tend to dislike this OS: if left uncared of for a while or not reformatted regularly, it becomes so helplessly slowed down and bloated by spyware it becomes a pain to use. I just can't get to an OS, having not started for two weeks, nagging me for an hour about updates to be made on all its mandatory security software. It's just not as streamlined as Linux or Mac.

A notable lab demonstrator of mine claimed that "Mac allows for more tasks to be done because it's beautiful, in and out". At first I was vaguely amused by this seemingly childish remark (before I got my own), but it turned out to be true. Beauty through simplicity, and efficiency through (apparent) simplicity. How well applications articulate with one another. How easily I can find a lost folder in the forest of them, in any dialog box. True, Linux didn't allow for such an integration, but at least laid down similar fundamentals: "one app for one task", as opposed to the one app does-it-all, dominant on Windows. Of course him and I were both very apt Windows users and dutifully used it to retrieve data, though I can't tell how much of our efficiency was due to the fact we knew our ways extensively in the OS.


Macs are less popular than PCs for several reasons, not necessarily related to cost-effectives. However, the PC has certainly a better cost-effectiveness ratio in the enterprise.
Pure cost is surely a factor. However, if enterprises would more often measure TCO, they would find out that having a large deployed Windows base can be more expensive, the majority of the cost not being the hardware, but support time. Of course it depends on your applications: large call centers often run a heavily locked-down version of Windows, protected by multiple levels of firewalls, with detailed custom security policies. In fact, Windows performs decently in such a protected environment.

This IT support friend of mine jokingly refers to it as the "work conspiracy", that Windows is dominant in the enterprise because it allows more of his peers to be employed (he justified his stance by referring to a colleague's job responsible for managing a pharmacy faculty running 100% Macs, about 300 machines, mixed students and staff: he always finds an hour to come by, have a chat and leave work at 3PM without any backlog, while colleagues responsible for the about 150 PCs in the library regularly work overtime, all 3 of them). He also added that "Windows is a communist OS: only compatible with itself". Even his position of mostly supporting student's machines (40% Macs), he only sees one Mac a day, vs. 10 to 15 PCs.

I was not aware of NeoOffice for iPad. Is it not available anymore?
They were in a difficult financial position due to NeoOffice requiring a significant amount of time to be integrated with the Mac (a void I always felt reduced the perceived value of the software for Mac users, since it's clearly made to emulate MS Office, instead of trying to pursue better integration with non-Windows), and they shut down the service years ago.

It may be true. However, it probably does not matter for the average user. People do not want recognized standards. People want to send a file and have this file read and understood by other people. That is what standards should be about.
Maybe I'm being idealistic (as many grad students are), but I believe that a recognized standard will be readable by other people, precisely because it would be software-agnostic. Of course, this is not the case, as MS Office proves daily. To mitigate it, I send PDFs. This is a truly universal format (although Windows XP couldn't read it natively and that Windows 8 support for it is very buggy).

Since Office Open XML documents can't be read perfectly without a recent MS Office, I don't consider it "readable and understandable by other people". There's a $150 tax barrier before readability. Only if they explicitly and affirmatively answer on their possession of MS Office I do send them an Office Open XML version.

I had a big problem in the past with Microsoft Word 2003. I was writing my Master's thesis and, just two months before the deadline, Word crashed and suddenly my 250-page .DOC was corrupt. Word crashed every time I opened the file. I tried to save another file. I tried to copy the content to another Word file. It did not work. Word crashed again. And again. I tried everything. And Word kept crashing. Panic.

What saved me was OpenOffice.org (it was called so back then). I tried to open the file in OpenOffice.org and it worked without crashing! Then I used OpenOffice.org to write the rest of my thesis. But the experience was not the best: the interface could be much improved and I missed the functionality of Word. Still, OpenOffice.org saved my life.

I moved to Word 2007 when it launched and it was certainly better.
That's one of the reasons I kept such an unreasonable amount of backups. OpenOffice isn't free of corruption, but since its format is, precisely, open source, one can always manually tweak the file and recover its content. That's not possible with Microsoft's binary formats, unless you pay big bucks for recovery software that may or may not yield acceptable results. A prime example of it is a proprietary data recovery tool I had to use to restore an ExFAT-formatted drive that another recovery software completely destroyed, ironically enough. Yet the newer one could retrieve files, but not actually rebuild the partition table. Luckily I haven't paid for it, but it serves as a strong reminder that one can't trust proprietary software, especially if no full-featured demo is available.

This reminds me of the story of a PhD student who, astonishingly kept only ONE copy of her thesis on one flimsy USB key, and started to have problems at the end of her 4th year. Four years writing, researching, saving, and just one USB key to save it all. Crazy is the unconsciousness of some deemed educated people.

Alas, I have to agree that OpenOffice interface isn't the best. Although very logical, it is "boring" and "ugly", according to my demonstrator's viewpoint. Since I left MS Office before the ribbon interface was made standard, I can't speak about its stability. I was told it's better, but daily crashes on university-controlled, standardized workstations with very common software is at least worrying.

I use .DOCX (OOXML). I know Microsoft did not implement it entirely when launched Office 2007, and I guess it only did with Office 2013.
All of which would reinforce the hypothesis they built a bloated specification on-purpose to compete with ODF.

However, I must use OOXML. Every publisher in my area accepts DOCX, and most of them only accept DOCX. Some of them accept other file formats, but I don't want to limit the number of journals that will consider my paper. So, I cannot just save in other format such as ODF or whatever I want. I must use DOCX.
That's what I call the "standardized exit point". You can use whatever process best fits your workflow, given that publication doesn't come that often (alas).

Yes, it is possible to stick with open source software, if you have to, or if you want to make this choice. I don't. I happily pay for software when I think it is worth it.
Being a grad student is an awkward position. Even though the university offered a free license for both MS Office and EndNote, I recalled the previous versions' bugginess and refused to go down that road. In the end it turned out to be a choice I made. I think developers should be compensated in some way for their work, and as I already said, I have nothing against paid software. I do, however, against proprietary formats that lock your choices down. Choosing a software should be based on actual capabilities, rather than out of habit and to be able to open non-standard formats. QtiPlot is a rare example of such software: source is open and you can actually build it if you're an advanced hacker. For the rest of us, there's a reasonably-priced option also providing personalized support. Also think of CodeWeaver. Reasonably priced, built on Wine, and polished for a perfect experience. But at the time, the university wasn't providing a license for SPSS, which is extremely expensive, even for students, and I needed to read a few SAV files. Luckily this part of the analysis could be done in PSPP just fine. Proprietary-format, reverse-engineered enough to be usable.

I consider most grad students except the luckiest to be a very poor species who can't afford most commercial software (even with education pricing), and precious money should rather be invested on solid hardware than on software without proper support (Microsoft doesn't offer support on their products AFAIK). I would have paid for NeoOffice if OpenOffice hadn't improved that much on the Mac. But the added benefits were too small to justify it, and I didn't like the 1-year license limitation principle. I did pay for Antidote, which, despite having no known competition, is very well designed despite its rather high $150 cost, and is tri-platform (and they have an iPad version as well), and comes with 3 licenses for use on separate machines.

However I wouldn't trust a non-open-sourced security software. Despite many praising it, I refused to buy 1Password as it can't undergo independent testing. If it happens to have a critical flaw, not just one but all passwords would be compromised. The passwords problem is still unsolved as far as I'm concerned. So I maintain a list of secure but memorable passwords I change regularly on different services. On the other hand, I do trust TrueCrypt; it was recently tested and found to have no serious vulnerabilities. Even not performing at all as designed, I trust ownCloud more than any hypothetical and proprietary cloud-managing software.


Wow. What a story.
Yeah, one showing why you should never skimp on hardware quality, yet never trust it, nor anyone.

In this case, it makes sense to use software that saves in filetypes that can be opened in as many systems as possible. Still, it is always possible to use OOXML and keep compatible.
In another department, all their databases where built with FileMaker. Not standard, but professors routinely handed pirated copies of the software. It was a non-issue for them.

However OOXML is currently not perfectly compatible except among post-2007 MS Office versions. And the issue was even worse when I was writing (See "format war" section).

I want to try and use software which was designed for the Mac so I can have the whole experience and not limit myself to cross-platform alternatives. The Mac is an expensive machine, and it wouldn't make sense to have it just to use the very same software I would be using on a PC.
Makes sense for me, however I expect full support to be provided with it, which is not the case of Microsoft's software. I considered the Mac to be premium hardware with superior integration and support, and paid the premium for it. Admittedly I haven't looked at specific software, except ones stemming from Linux. However there are a few, major software in my fields that are "Mac-compatible" but can't be decently called "designed for Mac". SPSS is a nightmare to install with all its Java inside, same goes for Matlab. Both are fiddly, heavy, and just plain fugly, as MS Office was (is?).

However, I want to keep compatibility. I have Microsoft Office 365 and I must say Office for Mac is much worse than Office for Windows. I can deviate from Office, but I must use software that will allow me to save in a format that Microsoft Office can open. I will do my last revision of my papers in Word. Always.
Funny. Most people I see buying Office 365 in store have problems installing in on up-to-date Windows 7 and Windows 8 PCs. It takes them three trials to finally have their key accepted, for some reason.

In science I considered perrenity of the data and its legibility to be more important than MS Office compatibility. Even as it was not a requirement in my project, I felt more comfortable knowing all of my data would be readable for decades. Microsoft could decide to cut support for previous formats, and nobody would have a say in that decision.

It may be true that proprietary software may want to keep you locked. If the software is gone, then you will probably have to re-make it all over again.

However, I must make two points here:

(1) If the proprietary software is wildly popular, then another software (either proprietary and/or open source) will probably provide some sort of functionality to import the files into the new application. This would allow the new software to get more users.

(2) You also run risks with open source software. I have seen several times the developer of open source software cease development. If someone in the community wants, he may keep the project alive, but will he? I have seen several open source software be lost in time because of this.
Well I can't agree much on point 1. MS Office and FileMaker are two extremely popular software, yet no other software can be considered serious competition when editing files created with the formers. Sorry, as much as I appreciate LibreOffice, their OOXML compatibility is not up to par. Not their fault, but still true.

On point 2, sure open-source developers do abandon their projects for various reasons, but usually a popular project has many followers, including developers ready to keep on updating it. For a short while, the blog engine I use, which is extremely popular for pure blogging (as opposed to general CMS), was considered to be EOLed, and I though about alternatives. But doubt didn't last long as seasoned developers quickly reassured the community they would keep on updating this engine, more secure and faster than Wordpress. I also felt not moving was the right decision, knowing the 3rd largest web hosting company still supports it.
Yes, it is all true.
As said, you may not recognize your needs here if you're only dealing with databases, reports and publications.

Sente will be smart to follow citation guidelines (e.g., changing from “Scholz, Smith, and Mitchell 2013″ to “Scholz et al. 2013″ after the first citation), and you can select what citation style you want to apply from an impressively long list of built-in formats (you can also modify/create your own style in the Bibliography Formats editor).

But hey, what if I have special needs when citing papers, such as adding page numbers (“Scholz 2011, p. 397″) or when I only cite the year (“As mentioned by Scholz (2011)”)? No worries, Sente has you covered! You can add modifiers (e.g., Scholz 2011@397, %Scholz 2011) to your citations, and Sente will adapt the way in which it displays the citation. You can find out more about these modifiers in the Sente manual under “Modifying In-Text Citations” (p. 261).
Huh, this isn't smart. It's the bare minimum expected of a reference manager software.
 
Even in one's non-native language! I learnt to conserve my energies for worthwhile subjects, and this is one.

Nice. Another long post. :D

What is your native language, by the way?

I've also read that, for some strange fiscal reason, Apple products were heavily taxed on import, but not necessarily common PCs. Even recently iPod Touch has been almost twice as expensive in Brazil than the rest of the world. Can you share more information about this?

Yes, sure.

In fact, it is not just Apple products. Brazil charges very heavy taxes on all electronic devices, produced by any company, and not just Apple. Steve Jobs once said Brazil had "crazy taxes".

Brazil always had very high taxes on imports. The historical explanation is that the government wanted to create barriers for imports in order to develop the national industry. More than one century later, the government keeps using this excuse, despite the industry having not developed the way it should. Taxes are very high for all products, and imports tax must be high in order for local products to be competitive. The high taxes finance a very inflated and inefficient government.

But taxes are high for all computers. This is why Apple (and perhaps Sony as well) is the only company that sells high-end laptops in Brazil. Apple laptops are not even the most expensive ones. Anyway, most laptops are not even for sale here and the offers are usually very low-end, and are still equipped with 3rd generation Intel Core processors. We do not have high-end laptops such as the Samsung Ativ Book 9 Plus, the Lenovo Yoga 2 Pro, the Acer Aspire S7, or the Asus Zenbook Infinity.

Look at this brief comparison (the prices are approximate, considering an average exchange rate of US$ 1 = R$ 2.20):

• iPad Air (16 GB, wi-fi): US$ 499 in the US, US$ 800 in Brazil
• 13-inch MacBook Air (4 GB, 128 GB): US$ 999 in the US, US$ 2,300 in Brazil
• 13-inch MacBook Pro with Retina Display (8 GB, 256 GB): US$ 1,499 in the US, US$ 3,500 in Brazil
• 15-inch MacBook Pro with Retina Display (8 GB, 256 GB): US$ 1,999 in the US, US$ 4,500 in Brazil
• 27-inch iMac (3.2 GHz version): US$ 1,799 in the US, US$ 4,300 in Brazil
• Dell XPS 12 (4 GB, 128 GB): US$ 1,199 in the US, US$ 2,600 in Brazil
• Dell Venue 11 Pro (64 GB): US$ 499 in the US, US$ 1,050 in Brazil
• Sony Vaio Pro (Core i7-4500U, 8 GB): US$ 1,799 in the US (with 256 GB), US$ 3,200 in Brazil (with 128 GB only)

And the list goes on and on. I have just obtained these numbers from the websites of these companies, so they are updated. I remember that the 13-inch Lenovo IdeaPad Yoga (the original model with 128 GB SSD) went on sale here for US$ 4,500, despite costing only US$ 999 in the US, which is absolutely insane. So, Macs are very expensive here, but just because every piece of hardware is very expensive here as well. It's not just Macs. All premium laptops sell for a very high price.

Yes, there are cheaper laptops around here. There are local companies that sell very low-end crappy plastic machines. If you think a Dell or an HP is crappy, think again. These machines are real crap, are equipped with obsolute processors (3rd generation Intel Core and sometimes even 2nd generation or Pentium or even Celeron), slow HDs, 1366x768 washed-out screens, bad keyboards and trackpads and loose parts. They are of course cheaper but they have a very low quality. However, even cheaper, they are still very expensive for what they are, and it would be possible to get one of these machines for as low as US$ 300 in the US. I cannot even think of getting one of these.

You probably heard people saying Macs were far more expensive here because people simply don't buy good computers (which, by the way, are hardly available outside of Apple's offers).

I used Windows daily from 1999 (first computer ever) to 2006, then Ubuntu from 2006 to 2008, then OS X from 2008, though not exclusively.

I have tried Ubuntu several times, but could not get used to it at all.

Can't deny I would never have to use it again, too, especially if the difference with common PCs was so high here as in Brazil. But here, an entry-level Mac can be had for $1500 after taxes, while a decent PC costs $1000. The difference wasn't great enough to justify lost productivity on inferior software / hardware / service, especially on something as critical as a thesis. Not to say I don't have a PC, I do, an older LG with about the same power as a 2009 MBP (same parts inside). I just don't use it because it feels so old and slow, even compared to my white MB, not mentioning noisy. It sits in a drawer most of the time, plus I don't think it would support Windows 7 properly, hardware-wise. I can be mistaken, but coming from 8 years of dominant Unix-likes use, the transition to Windows would feel like a big, backward step. In fact, I call all my Windows installations "legacy". This pretty much sums it. And a PC can run Linux of course, so I don't feel tied by Windows. The only moment I still use it: on the job, when the Macs are taken; runs on a buggy machine, full of spyware, and dated hardware. Maybe that's why I tend to dislike this OS: if left uncared of for a while or not reformatted regularly, it becomes so helplessly slowed down and bloated by spyware it becomes a pain to use. I just can't get to an OS, having not started for two weeks, nagging me for an hour about updates to be made on all its mandatory security software. It's just not as streamlined as Linux or Mac.

A notable lab demonstrator of mine claimed that "Mac allows for more tasks to be done because it's beautiful, in and out". At first I was vaguely amused by this seemingly childish remark (before I got my own), but it turned out to be true. Beauty through simplicity, and efficiency through (apparent) simplicity. How well applications articulate with one another. How easily I can find a lost folder in the forest of them, in any dialog box. True, Linux didn't allow for such an integration, but at least laid down similar fundamentals: "one app for one task", as opposed to the one app does-it-all, dominant on Windows. Of course him and I were both very apt Windows users and dutifully used it to retrieve data, though I can't tell how much of our efficiency was due to the fact we knew our ways extensively in the OS.

Nice. After trying my first Mac, well, it was just great. I keep going to Windows and coming back. I use Windows at work, and a Mac at home (and sometimes Parallels or BootCamp).

Pure cost is surely a factor. However, if enterprises would more often measure TCO, they would find out that having a large deployed Windows base can be more expensive, the majority of the cost not being the hardware, but support time. Of course it depends on your applications: large call centers often run a heavily locked-down version of Windows, protected by multiple levels of firewalls, with detailed custom security policies. In fact, Windows performs decently in such a protected environment.

This IT support friend of mine jokingly refers to it as the "work conspiracy", that Windows is dominant in the enterprise because it allows more of his peers to be employed (he justified his stance by referring to a colleague's job responsible for managing a pharmacy faculty running 100% Macs, about 300 machines, mixed students and staff: he always finds an hour to come by, have a chat and leave work at 3PM without any backlog, while colleagues responsible for the about 150 PCs in the library regularly work overtime, all 3 of them). He also added that "Windows is a communist OS: only compatible with itself". Even his position of mostly supporting student's machines (40% Macs), he only sees one Mac a day, vs. 10 to 15 PCs.

Well, I guess it depends. There is more people qualified to render support services on Windows than on Macs. I guess Mac support is more expensive. Here in Brazil, Mac support for the enterprise is simply non-existent.

They were in a difficult financial position due to NeoOffice requiring a significant amount of time to be integrated with the Mac (a void I always felt reduced the perceived value of the software for Mac users, since it's clearly made to emulate MS Office, instead of trying to pursue better integration with non-Windows), and they shut down the service years ago.

Oh, I see.

Maybe I'm being idealistic (as many grad students are), but I believe that a recognized standard will be readable by other people, precisely because it would be software-agnostic. Of course, this is not the case, as MS Office proves daily. To mitigate it, I send PDFs. This is a truly universal format (although Windows XP couldn't read it natively and that Windows 8 support for it is very buggy).

Since Office Open XML documents can't be read perfectly without a recent MS Office, I don't consider it "readable and understandable by other people". There's a $150 tax barrier before readability. Only if they explicitly and affirmatively answer on their possession of MS Office I do send them an Office Open XML version.

Actually, as everybody is supposed to have Word installed, then it should be fine to use Word file format. Strange as it may sound, OOXML is more "standardized" and more universally understood than ODT. If I send an OOXML file to a list of 10 random friends, all of them will be able to open the file. However, if I send an ODT file to the very same list, perhaps only one (or none) will be able to open it (or figure out how to open it).

That's one of the reasons I kept such an unreasonable amount of backups. OpenOffice isn't free of corruption, but since its format is, precisely, open source, one can always manually tweak the file and recover its content. That's not possible with Microsoft's binary formats, unless you pay big bucks for recovery software that may or may not yield acceptable results. A prime example of it is a proprietary data recovery tool I had to use to restore an ExFAT-formatted drive that another recovery software completely destroyed, ironically enough. Yet the newer one could retrieve files, but not actually rebuild the partition table. Luckily I haven't paid for it, but it serves as a strong reminder that one can't trust proprietary software, especially if no full-featured demo is available.

You should be right, probably. However, as far as I am concerned, OOXML is much better in this regard, and similar to ODT in many ways.

This reminds me of the story of a PhD student who, astonishingly kept only ONE copy of her thesis on one flimsy USB key, and started to have problems at the end of her 4th year. Four years writing, researching, saving, and just one USB key to save it all. Crazy is the unconsciousness of some deemed educated people.

Yes. I learned my lesson now.

Alas, I have to agree that OpenOffice interface isn't the best. Although very logical, it is "boring" and "ugly", according to my demonstrator's viewpoint. Since I left MS Office before the ribbon interface was made standard, I can't speak about its stability. I was told it's better, but daily crashes on university-controlled, standardized workstations with very common software is at least worrying.

Actually, I don't care about it being boring or ugly. I care about usability. Clearly, "ergonomics" is not a concept that populates developers of open source software.

As your post is too long, I will stop here for a while. I will return later. :D
 
I've written papers shorter thant he average post length in this thread :p

I can imagine. :eek:

By the way, you academics should know that Scrivener is for sale today at Mac Update (50% off), for US$ 22.50. You can see it at http://www.mupromo.com. Mac Update has not featured many worth discounts lately. However, Scrivener is software worth getting and I think it is worth it.
 
Last edited:
I can imagine. :eek:

By the way, you academics should know that Scrivener is for sale today at Mac Update (50% off), for US$ 22.50. You can see it at http://www.mupromo.com. Mac Update has not featured many worth discounts lately. However, Scrivener is software worth getting and I think it is worth it.

Have had it for a couple of years now but never really found a reason to migrate to it form Word.

Have had it for a couple of years now but never really found a reason to migrate to it form Word. Does it play nice with reference manager, particularly Zotero & Mendeley ?
 
Last edited:
Have had it for a couple of years now but never really found a reason to migrate to it form Word.

Have had it for a couple of years now but never really found a reason to migrate to it form Word. Does it play nice with reference manager, particularly Zotero & Mendeley ?

Actually, I am just beginning to use it. I had never widely used it before, but I am starting too, as it looks nice and I am getting tired of waiting for Microsoft deliver me a better Word for Mac.

I do not know if it plays well with Zotero or Mendeley. It is supposed to work with reference managers, but I have not yet tried it widely. Zotero and Mendeley are both cross-platform, and they are not expected to work with word processors designed for Mac (even though there is a version of Scrivener for Windows). They work well with Word and LibreOffice Writer, but I would not expect it to work with Scrivener.

In fact, I found this on Scrivener and Zotero integration: http://dahl.at/wordpress/2014/02/05/scrivener-and-zotero/

As you can see, they do not integrate too well, no cite-while-you-write function, but you can work with both.

I can tell you that Scrivener works with Sente, as Sente recognizes it as a word processor (but I guess it will only scan files after you compile the draft). And I think Papers also works with Scrivener (including the Magic Citations feature). Apart from that, I don't think there would be much integration between Sente and reference managers.

Anyway, if you don't already have Scrivener 2 nor the right to upgrade, you can buy the upgade for US$ 25 if you ever think of switching to it. You don't need this promo, so you can decide to move to Scrivener any time you want, if you want it. :D
 
Nice. Another long post. :D

What is your native language, by the way?
French.

Yes, sure.

In fact, it is not just Apple products. Brazil charges very heavy taxes on all electronic devices, produced by any company, and not just Apple. Steve Jobs once said Brazil had "crazy taxes".

Brazil always had very high taxes on imports. The historical explanation is that the government wanted to create barriers for imports in order to develop the national industry. More than one century later, the government keeps using this excuse, despite the industry having not developed the way it should. Taxes are very high for all products, and imports tax must be high in order for local products to be competitive. The high taxes finance a very inflated and inefficient government.
Different countries, same problem.
But taxes are high for all computers. This is why Apple (and perhaps Sony as well) is the only company that sells high-end laptops in Brazil. Apple laptops are not even the most expensive ones. Anyway, most laptops are not even for sale here and the offers are usually very low-end, and are still equipped with 3rd generation Intel Core processors. We do not have high-end laptops such as the Samsung Ativ Book 9 Plus, the Lenovo Yoga 2 Pro, the Acer Aspire S7, or the Asus Zenbook Infinity.

Look at this brief comparison (the prices are approximate, considering an average exchange rate of US$ 1 = R$ 2.20):

• iPad Air (16 GB, wi-fi): US$ 499 in the US, US$ 800 in Brazil
• 13-inch MacBook Air (4 GB, 128 GB): US$ 999 in the US, US$ 2,300 in Brazil
• 13-inch MacBook Pro with Retina Display (8 GB, 256 GB): US$ 1,499 in the US, US$ 3,500 in Brazil
• 15-inch MacBook Pro with Retina Display (8 GB, 256 GB): US$ 1,999 in the US, US$ 4,500 in Brazil
• 27-inch iMac (3.2 GHz version): US$ 1,799 in the US, US$ 4,300 in Brazil
• Dell XPS 12 (4 GB, 128 GB): US$ 1,199 in the US, US$ 2,600 in Brazil
• Dell Venue 11 Pro (64 GB): US$ 499 in the US, US$ 1,050 in Brazil
• Sony Vaio Pro (Core i7-4500U, 8 GB): US$ 1,799 in the US (with 256 GB), US$ 3,200 in Brazil (with 128 GB only)

And the list goes on and on. I have just obtained these numbers from the websites of these companies, so they are updated. I remember that the 13-inch Lenovo IdeaPad Yoga (the original model with 128 GB SSD) went on sale here for US$ 4,500, despite costing only US$ 999 in the US, which is absolutely insane. So, Macs are very expensive here, but just because every piece of hardware is very expensive here as well. It's not just Macs. All premium laptops sell for a very high price.

Yes, there are cheaper laptops around here. There are local companies that sell very low-end crappy plastic machines. If you think a Dell or an HP is crappy, think again. These machines are real crap, are equipped with obsolute processors (3rd generation Intel Core and sometimes even 2nd generation or Pentium or even Celeron), slow HDs, 1366x768 washed-out screens, bad keyboards and trackpads and loose parts. They are of course cheaper but they have a very low quality. However, even cheaper, they are still very expensive for what they are, and it would be possible to get one of these machines for as low as US$ 300 in the US. I cannot even think of getting one of these.
It's awful how your description is strikingly similar to our own situation…With imported cheese! Even imported, with added shipping fees, they would normally end up much less expensive than local ones, which have to work with extremely high production cost due to milk price not being set by market alone, but with a minimum price and artificially-limited supply. Even milk-producers can't do what seems logical, namely milk the animal, put it in bins, and start making different cheeses. They have to sell it to a single, mandatory customer then buy it back at a greatly inflated price before processing it.

So to avoid disloyal competition, the government has set up extremely high taxes on imported cheese so they still cost a tad more than locally-produced ones. Even buying it abroad and packing it properly for non-regrigerated air travel, you still have to pay a hefty tax if above 1Kg. No wonder some mafiosi families have entered the milk business.

However when comparing prices, you also have to match with the minimum wage, pretty much the only one a student can afford to get here. These prices wouldn't seem that big if minimum wage was, say, $20. I was told by a student returning from an exchange with Brazil that tuition and living fees for foreigners are nowhere comparable to those of the US. So there aren't just bad sides of living in Brazil: you just have to buy your computer abroad while on a trip.

You probably heard people saying Macs were far more expensive here because people simply don't buy good computers (which, by the way, are hardly available outside of Apple's offers).
Makes sense.

I have tried Ubuntu several times, but could not get used to it at all.
Weird. It's so similar to OS X.

Nice. After trying my first Mac, well, it was just great. I keep going to Windows and coming back. I use Windows at work, and a Mac at home (and sometimes Parallels or BootCamp).
I try to stay away from Windows most of the time. I can be considered privileged enough to have a workplace where Mac is dominant (and not an art-related business) and haul my MBP everywhere I go. Should the need arise, I can always bring an external HDD with Windows VirtualBox installation. But never found a valid reason to shut my Mac down to boot Windows in BootCamp, or waste valuable internal storage. Yet again, I couldn't find a compelling reason to get Parallels or other proprietary virtualizing software compared to the only open-source alternative I know. Yes, monopoly can exist in the open-source world, and as everywhere else, lack of competition kills innovation. The same could be said of LibreOffice, as other open-source software able to open ODF documents do exist, but with a fraction of the functionality

Well, I guess it depends. There is more people qualified to render support services on Windows than on Macs. I guess Mac support is more expensive. Here in Brazil, Mac support for the enterprise is simply non-existent.
I wouldn't be able to say. Maybe things stay equal since Mac is very seldom used in enterprises large enough to require a dedicated Mac IT guy that would be more than a Windows-er with some Mac experience.

Actually, as everybody is supposed to have Word installed, then it should be fine to use Word file format. Strange as it may sound, OOXML is more "standardized" and more universally understood than ODT. If I send an OOXML file to a list of 10 random friends, all of them will be able to open the file. However, if I send an ODT file to the very same list, perhaps only one (or none) will be able to open it (or figure out how to open it).
Not "everybody is supposed to have Word installed". It doesn't come standard with most computers purchased, even though post-XP Windows may be able to view it. As far as the "open data" movement goes in the EU, governments are required to make data available in a format citizens can read without purchasing anything. As a good sharing habit, should I need to send a document to someone, I will always ask them if they have a specific piece of software installed. Otherwise, I send a PDF. Luckily LibreOffice is able to open a majority of OOXML without a fuss.

If all your friends can open an OOXML, it indeed implies they have MS Office installed. Of course pirating MS Office is always possible and would nullify the "$150 tax barrier"; I simply wanted to stay away from bad habits I had in Windows of pirating pretty much every piece of useful software. I was deeply convinced an OS as advanced as OS X should only need a few, free applications to meet all my needs, and felt I would be denying UNIX's purported power, given that every standard was supported natively. Quite naïve.

On the other hand, recognizing an OOXML file but not an ODF one may have to do with Microsoft's pervasive and cunning marketing, going as far as licensing its viewability to third-parties or forcing them to use the mention "Brand recommends Windows" on top of their pages. While LibreOffice doesn't have such strategies in place. Much like I suspect most Chrome users haven't grew to such numbers because it's an inherently superior piece of software, but because of heavy-handed marketing from Google. Meanwhile I do consider Firefox to be better, valued for a flexibility Chrome doesn't have.

You should be right, probably. However, as far as I am concerned, OOXML is much better in this regard, and similar to ODT in many ways.
Can't say from personal experience, but a few times a week I do see people who have lost their OOXML file in an MS Office crash. Even stranger that it happens on a standardized installation of Windows 7 + MS Office, in a controlled environment, and served from a single point to many thin-client terminals.

Actually, I don't care about it being boring or ugly. I care about usability. Clearly, "ergonomics" is not a concept that populates developers of open source software.
You're right, alas. When I made remarks on lack of ergonomics in otherwise very good open-source software, I always was greeted by answers such as "If you don't like it, fork it!", "FOSS developers only develop for themselves.", "It's not un-ergonomic, it's just you don't know to use it", "why a GUI? Can't you type?", or the classical "RTFM".

Well I am strongly in favor of sensible GUIs, where logic and deduction leads a novice user to the right function, instead of relying on one's memory, or static, large man pages that need to be scrolled up and down in search of the exact syntax to be used. If OO / LibO attained such a level long ago, it still lacks a "beauty" factor, as per my demonstrator.

As your post is too long, I will stop here for a while. I will return later. :D
Waiting for you. Has been a pleasure :)

I do not know if it plays well with Zotero or Mendeley. It is supposed to work with reference managers, but I have not yet tried it widely. Zotero and Mendeley are both cross-platform, and they are not expected to work with word processors designed for Mac (even though there is a version of Scrivener for Windows). They work well with Word and LibreOffice Writer, but I would not expect it to work with Scrivener.
Zotero works perfectly with LibreOffice, and has been from the start. I don't know where you're getting you informations from. Of course it suffers from the same problem pointed earlier, namely unimaginative ergonomics.

That's a real problem of proprietary software: if you want to choose the best software for a given usage, nothing guarantees that you'll be able to interface it with your preferred software for another task. If, for example, you prefer Sente to manage your references, and Scrivener to write, you have to add a step, reducing your efficiency, or just throw more money at the issue for some developer to make something for you. Not that it doesn't exist in the FOSS world, but just one interested developer is enough for a project to start.

What is Scrivener much better at than a standard word processor, anyway?
 
I can imagine. :eek:

Lol....I just checked. You opening post is 831 words long and it is short by the standards of the thread. I have a published "Commentary" style manuscript that is only 656 words long and been cited several times - ha :p
 

Cool.

Different countries, same problem.
It's awful how your description is strikingly similar to our own situation…With imported cheese! Even imported, with added shipping fees, they would normally end up much less expensive than local ones, which have to work with extremely high production cost due to milk price not being set by market alone, but with a minimum price and artificially-limited supply. Even milk-producers can't do what seems logical, namely milk the animal, put it in bins, and start making different cheeses. They have to sell it to a single, mandatory customer then buy it back at a greatly inflated price before processing it.

So to avoid disloyal competition, the government has set up extremely high taxes on imported cheese so they still cost a tad more than locally-produced ones. Even buying it abroad and packing it properly for non-regrigerated air travel, you still have to pay a hefty tax if above 1Kg. No wonder some mafiosi families have entered the milk business.

However when comparing prices, you also have to match with the minimum wage, pretty much the only one a student can afford to get here. These prices wouldn't seem that big if minimum wage was, say, $20. I was told by a student returning from an exchange with Brazil that tuition and living fees for foreigners are nowhere comparable to those of the US. So there aren't just bad sides of living in Brazil: you just have to buy your computer abroad while on a trip.

Well, at least it appears that only cheese suffers from high import taxes in France.

In Brazil, we have high taxes for nearly every single product. Import taxes are high, but they are not the only taxes applicable to import products. In fact, there is a tax that is applicable over the price of the product, over the shipping costs, over other taxes and over itself. Every product is nearly double the price of products in the U.S., including cars, clothes and electronic devices.

One can buy a laptop abroad, but then he will have to go through the customs office. If one get caught with a laptop without declaring it, will have to pay a 50% import tax over whatever exceeds US$ 500 plus a fine of 50%. Yes, 100% over what exceeds US$ 500.

Weird. It's so similar to OS X.

Really? Are you serious?

OS X is very ergonomical. It is a pleasure to use. Ubuntu and other Linux distributions I tried were similar to OS X, but without the beauty, the ergonomics, the finesse, the joy of use, the user experience and everything else that makes Macs special.

I try to stay away from Windows most of the time. I can be considered privileged enough to have a workplace where Mac is dominant (and not an art-related business) and haul my MBP everywhere I go. Should the need arise, I can always bring an external HDD with Windows VirtualBox installation. But never found a valid reason to shut my Mac down to boot Windows in BootCamp, or waste valuable internal storage. Yet again, I couldn't find a compelling reason to get Parallels or other proprietary virtualizing software compared to the only open-source alternative I know. Yes, monopoly can exist in the open-source world, and as everywhere else, lack of competition kills innovation. The same could be said of LibreOffice, as other open-source software able to open ODF documents do exist, but with a fraction of the functionality

My only reason to use Windows has been Microsoft Office. Microsoft Office for Mac does not hold a candle to Office for Windows and, therefore, I use the Windows version. I have used Windows in the past for OCR (with is better on Windows) and for playing games. And now and then there are specific programs that only run on Windows. Apart from that, I don't think I need Windows.

I wouldn't be able to say. Maybe things stay equal since Mac is very seldom used in enterprises large enough to require a dedicated Mac IT guy that would be more than a Windows-er with some Mac experience.

That would make Mac support expensive. Supply and demand.

Not "everybody is supposed to have Word installed". It doesn't come standard with most computers purchased, even though post-XP Windows may be able to view it. As far as the "open data" movement goes in the EU, governments are required to make data available in a format citizens can read without purchasing anything. As a good sharing habit, should I need to send a document to someone, I will always ask them if they have a specific piece of software installed. Otherwise, I send a PDF. Luckily LibreOffice is able to open a majority of OOXML without a fuss.

Well, maybe not everybody. But in a corporate environment, if you don't have Word you're dead.

If all your friends can open an OOXML, it indeed implies they have MS Office installed. Of course pirating MS Office is always possible and would nullify the "$150 tax barrier"; I simply wanted to stay away from bad habits I had in Windows of pirating pretty much every piece of useful software. I was deeply convinced an OS as advanced as OS X should only need a few, free applications to meet all my needs, and felt I would be denying UNIX's purported power, given that every standard was supported natively. Quite naïve.

On the other hand, recognizing an OOXML file but not an ODF one may have to do with Microsoft's pervasive and cunning marketing, going as far as licensing its viewability to third-parties or forcing them to use the mention "Brand recommends Windows" on top of their pages. While LibreOffice doesn't have such strategies in place. Much like I suspect most Chrome users haven't grew to such numbers because it's an inherently superior piece of software, but because of heavy-handed marketing from Google. Meanwhile I do consider Firefox to be better, valued for a flexibility Chrome doesn't have.

I actually subscribe to Office 365 because I think it is worth it. Office is good software. Microsoft has done it right. Microsoft Office 2013 is perhaps the best single piece of software ever, and the most fully-featured. I cannot praise it enough. Too bad Microsoft won't make a decent version of it for Mac.

There are more reasons for Chrome becoming popular (and LibreOffice not) than Google's marketing:

1. Chrome is better than Internet Explorer. In fact, Explorer is bad software. It's no wonder why people seek alternatives even though there is a free web browser that comes installed on Windows.

2. LibreOffice is not as good as Microsoft Office, the interface is not as good, it is more inefficient to be used, and it is not as compatible.

3. Microsoft Office has its way into the enterprise, and, to play the game, one must have it installed, and not LibreOffice or other cheap alternative. For the corporate market, you must have the real thing.

Can't say from personal experience, but a few times a week I do see people who have lost their OOXML file in an MS Office crash. Even stranger that it happens on a standardized installation of Windows 7 + MS Office, in a controlled environment, and served from a single point to many thin-client terminals.

I have never heard of that. An OOXML file is a zipped file just as ODT. Try to open it with any ZIP utility.

You're right, alas. When I made remarks on lack of ergonomics in otherwise very good open-source software, I always was greeted by answers such as "If you don't like it, fork it!", "FOSS developers only develop for themselves.", "It's not un-ergonomic, it's just you don't know to use it", "why a GUI? Can't you type?", or the classical "RTFM".

Well I am strongly in favor of sensible GUIs, where logic and deduction leads a novice user to the right function, instead of relying on one's memory, or static, large man pages that need to be scrolled up and down in search of the exact syntax to be used. If OO / LibO attained such a level long ago, it still lacks a "beauty" factor, as per my demonstrator.

I just gave up those guys. That is one of the reasons why open source will not work. People are not interested in pleasing anyone.

I prefer to spend my time working so I can buy better software (and other things) instead of trying to figure out how to use free (as in beer) open source software.

Yes, there is good open source software. But they are exceptions. And LibreOffice is so fully-featured because it was actively developed as proprietary software before becoming open source.

Waiting for you. Has been a pleasure :)

I will do that. It's in my bucket list... LOL

Zotero works perfectly with LibreOffice, and has been from the start. I don't know where you're getting you informations from. Of course it suffers from the same problem pointed earlier, namely unimaginative ergonomics.

I was talking about Scrivener. Scrivener does not play well with Zotero apparently. Zotero does. In fact, I think Zotero is a good piece of software. I preferred it over Endnote. If I could go back in time, I would have used it for my PhD thesis instead of Endnote.

That's a real problem of proprietary software: if you want to choose the best software for a given usage, nothing guarantees that you'll be able to interface it with your preferred software for another task. If, for example, you prefer Sente to manage your references, and Scrivener to write, you have to add a step, reducing your efficiency, or just throw more money at the issue for some developer to make something for you. Not that it doesn't exist in the FOSS world, but just one interested developer is enough for a project to start.

Yes, you have a point. Still, some proprietary software operate well with others.

What is Scrivener much better at than a standard word processor, anyway?

Well, it has several features for writers and looks really powerful. I will figure out why it is so good and then I can tell you. But I have to use it widely first.
 
Well, at least it appears that only cheese suffers from high import taxes in France.
I'm not in France :D I would bet French cheeses are specifically aimed at with such a legislation, as they are a "poor's food" there, compared to a luxury here.

In Brazil, we have high taxes for nearly every single product. Import taxes are high, but they are not the only taxes applicable to import products. In fact, there is a tax that is applicable over the price of the product, over the shipping costs, over other taxes and over itself. Every product is nearly double the price of products in the U.S., including cars, clothes and electronic devices.

One can buy a laptop abroad, but then he will have to go through the customs office. If one get caught with a laptop without declaring it, will have to pay a 50% import tax over whatever exceeds US$ 500 plus a fine of 50%. Yes, 100% over what exceeds US$ 500.
I meant you could buy the laptop abroad, trash the packaging, and just declare is as part of your luggage. But I read once tax law in Brazil was extremely complicated, in addition to unfair.


Really? Are you serious?

OS X is very ergonomical. It is a pleasure to use. Ubuntu and other Linux distributions I tried were similar to OS X, but without the beauty, the ergonomics, the finesse, the joy of use, the user experience and everything else that makes Macs special.
I can't deny it, and referred to Ubuntu in its intention and core, but not its execution, while better than Windows XP, required too many fallback to the command line. I finally chose Mac because no Ubuntu desktop would be sold at a decent price compared to same laptops equipped with MS Windows ("The Microsoft tax penalty"), I refused to pay the Windows tax as it would come with Windows installed although I already had a separately-bought license (a hot topic on Linux forum is how to get refunded for that disguised "tax", which is unlawful according to many consumer's laws, but a de facto standard.), and that OS X had all the advantages of Linux, and added compatibility with commercial software should I need them. Top-rated service, software and hardware integration, and better overall engineering were bonuses.

My only reason to use Windows has been Microsoft Office. Microsoft Office for Mac does not hold a candle to Office for Windows and, therefore, I use the Windows version. I have used Windows in the past for OCR (with is better on Windows) and for playing games. And now and then there are specific programs that only run on Windows. Apart from that, I don't think I need Windows.
Agreed, MS Office is better on Windows, and I suspect Microsoft deliberately crippled the Mac version to push heavy users toward Windows.

Never had much success with OCR applications in Windows, though. A one-page typed text would always end up prompting 30-ish confirmations, with the suggestions wrong most of the time. One of my unsolved problems. I use Acrobat for that purpose on the Mac. What application have you used that yielded acceptable results?

That would make Mac support expensive. Supply and demand.
Not conditional, it is currently expensive because supply is low. But all things being equal, lower maintenance requirements on Macs would make the bill lower.

Well, maybe not everybody. But in a corporate environment, if you don't have Word you're dead.
Corporations and universities usually have a site license for MS Office. But given the current context of economic downtimes, these expensive site licenses have been put into question, as ditching Microsoft Office would result in millions in saved dollars. Our university has started to acknowledge the issue, and while they won't ditch MS Office anytime soon, at least LibreOffice is also installed so students aren't required to buy or pirate MS Office in order to be able to read and edit their files at home. I once wrote an extensive letter to the director of a national library he boasted aimed to help poorer classes access culture and computer tools. I argued that having only MS Windows + MS Office installed would not help disadvantaged masses to view computers as affordable, accessible machines, and that, by denying choice (a few Mac / Linux would have done the trick), denied democratization of computer knowledge, while normalizing piracy. It has to be understood that most used computers sellers here sell old machines at a high price, equipped entirely with pirated copies of Windows, MS Office being available on demand, all of it without proper keys. I think this is inherently wrong.

I actually subscribe to Office 365 because I think it is worth it. Office is good software. Microsoft has done it right. Microsoft Office 2013 is perhaps the best single piece of software ever, and the most fully-featured. I cannot praise it enough. Too bad Microsoft won't make a decent version of it for Mac.
I have seen way too many people buying the small box containing Office 365 license having problems just to install it on current Windows 7 and 8, burning all three licenses for just one successful install to call it "worth it". Even the IT guy gave up on helping them since Microsoft obviously did something wrong, but provide no channel to contact them. Instead, he asks students to go back where they bought the box, and have it installed in store.

In another software category, I subscribe to SpiderOak, not because of its open-sourced-ness, but because it's tri-platform and secure. I mean, I happily paid for it at first, but since a year and a half, it can't re-download uploaded files; some do come fine, some others stall, for reasons their own engineers can't understand. They gave me a free year of storage, added 30% storage space for the same price, and they proposed that I take a whole new account, but I refused since they wouldn't be able to transplant the database on their end, or even mail me a physical disk with all my encrypted data on it. I don't have anything against paying for service, if only working! I can't even say an open-source model would have solved it, since I wanted to test ownCloud on my hosting space because it boasted on being a valid iCloud replacement with an option to encrypt files on the servers, but in reality, it repeatedly fails outside advertised boundaries, i.e. not requiring any root access to be installed and configured.

There are more reasons for Chrome becoming popular (and LibreOffice not) than Google's marketing:

1. Chrome is better than Internet Explorer. In fact, Explorer is bad software. It's no wonder why people seek alternatives even though there is a free web browser that comes installed on Windows.
We all agree on that. However there has been a long-time alternative to IExplore that is mature, flexible, and respects the user's privacy: Firefox. However it doesn't have any marketing strategy, and relies on word-of-mouth positive reviews.

2. LibreOffice is not as good as Microsoft Office, the interface is not as good, it is more inefficient to be used, and it is not as compatible.
Let's be fair, it is claimed to do "80 to 90% of what MS Office does", the remaining 10 to 20% being mostly advanced functions we may or may not use. My giant spreadsheet was clearly in that exception range, but as very few pushed it so far, it was mostly unheard of, although way below the theoretical maximums allowed. I think there is a bug somewhere, because it can still easily crash on a machine with 16GiB RAM before saturating it. If I had to grow such a file again, I would probably choose Excel and add a CSV exporting step, or try to find a proper bidirectional ODS converter (to my knowledge, Gnumeric has its own format (!), and supports ODS as well). Added manual work, but probably better efficiency overall. However, given MS Office nasty tendency of modifying files and formatting automatically against a user's will, I wouldn't trust it too much. For that reason, while I would go with Excel for performance reasons, I would stay with Writer for the thesis; neither it nor Word are very good integrating images in the text, and perhaps only Pages does it right, at the expense of non-existent reference managers support.

This is probably my biggest gripe against MS Office: too many bugs disguised as functionality, and just like Windows, settings are difficult to find, not mentioning non-existent support.

3. Microsoft Office has its way into the enterprise, and, to play the game, one must have it installed, and not LibreOffice or other cheap alternative. For the corporate market, you must have the real thing.
If a corporation wants me to use MS Office, it will have to provide it. Otherwise this amounts to pay for the right to work. LibreOffice isn't a "cheap" alternative, it is a full-fledged office suite that may or may not meet one's needs, free of charge, and with a large community support. Not to say I praise it (in fact I am still waiting for an answer on proper ownCloud alternatives – doesn't behave as described: "you're doing it wrong" was I told –, and a configuration guide for OpenVPN), but at least in this case it's efficient. Note that also applies to any job requiring the applicant to have a specific tool, without compensating properly for the added expense, e.g. common "delivery man with car needed", and no mention of a per-km paid expense.


I have never heard of that. An OOXML file is a zipped file just as ODT. Try to open it with any ZIP utility.
One would have to find it first. Default MS Office settings, just like Windows, can be customized to infinity, and no installation is identical to the next one, especially in a large installed base of Microsoft technologies. This probably contributes to the high manpower required to maintain such networks.

I just gave up those guys. That is one of the reasons why open source will not work. People are not interested in pleasing anyone.
Sure. But the same could be said of software companies that completely dominate the market: whatever their users are thinking, it's not like they are going to go anywhere soon since they would still pay for the software in the form of bundled sales, and that hardware manufacturers aren't free since cutting ties with Microsoft would result in higher prices due to M$ not paying part of their marketing expenses. That leads me to say most people don't "choose" Windows. They rather "default" to Microsoft buying a price that suits them, without consideration for future maintenance costs. I strongly advocate that the laws forbidding bundled sales should be enforced, and that a given computer be presented with a choice of OSes, separately listing the license price for proprietary software, unless the same manufacturer did both hardware and software. Sure many would still choose Windows, but at least choice would be provided, much like the legendary Windows version that asked users which browser they wanted to install first. I never came across this version, though. At least there would be competition.

A minority of open-source developers do care about the user experience, and mostly develop on OS X. Oh well.
I prefer to spend my time working so I can buy better software (and other things) instead of trying to figure out how to use free (as in beer) open source software.
Well as I said I haven't got anything against paid-for software. RedHat is in fact a company that makes money providing premium support for its software, while leaving the source open for others to use, calling it Fedora. Ubuntu was criticized by some purists because it originates from a for-profit company, Canonical, that imposed its design decisions (M. Shuttleworth once declared that "every software that requires the use of a command-line on Ubuntu is a bug that needs to be tackled". Some were outraged, I was in bliss). I feel this is simply stupid, as everything is still open. Even the Linux kernel isn't democratically-developped.

I do have, however, a fear of locking myself out of my own files because the format wouldn't be interoperable, and the software editor may choose to cut support for older versions, even remotely. The same reason was invoked when the EU wanted to pass into law that all government-issued files had to be available in open formats. This was also an argument in the StopPalladium campaign (early 2000s) against the the TPM to be integrated in future hardware, with the rumor that Microsoft would use it in its Windows to arbitrarily lock access to a user's own files. In the end, this opaque chip is now integrated in most computers, but editors so far haven't made the jump to use its full capabilities.

Note this isn't specific to Microsoft: Rosetta, as a piece of proprietary software, was never released as open source for developers to build upon since its license allegedly forbade it. The few PowerPC proprietary software remaining are the other half of the problem, as releasing the source would at least allow for continued functionality. Let's not forget that Darwin, OS X's core, is also open-source. Maybe part-time developers haven't build anything on top of it because OS X was so good from the start :D As you point, this may be the right road to follow:
Yes, there is good open source software. But they are exceptions. And LibreOffice is so fully-featured because it was actively developed as proprietary software before becoming open source.
In fact, given all these examples, I'd rather have development of both FOSS and proprietary software revolve around the use of common, open formats: ODF is criticized for allowing to many unspecified elements (probably why Gnumeric can use it, and Calc just not interpreting these elements), and OOXML to be intentionally bloated to prevent anyone else than Microsoft to implement it properly.


I was talking about Scrivener. Scrivener does not play well with Zotero apparently. Zotero does.
:confused:

In fact, I think Zotero is a good piece of software. I preferred it over Endnote. If I could go back in time, I would have used it for my PhD thesis instead of Endnote.
AFAIK you would have to redo all your bibliography again since EndNote refuses to use open standards for storage. That is one tough choice I had to make when choosing a reference manager: while I was sure I didn't want EndNote (chiefly because there was no LibreOffice compatibility), I wasn't so sure about the others since, while open-source, they wouldn't necessarily use the same formats, nor interact with one another. So I more or less defaulted to Zotero as it was available on all platforms, and that Mendeley, while more ergonomic, lacked compatibility with LibreOffice.


Yes, you have a point. Still, some proprietary software operate well with others.
Antidote is an example of such a software. Proprietary, but respecting users' choice, although support isn't that good.
Well, it has several features for writers and looks really powerful. I will figure out why it is so good and then I can tell you. But I have to use it widely first.
Well I wouldn't pay premium price for a software I am not even sure would be useful, as I'm not a writer. It wouldn't be wise. But a fully-featured version that would include a "Please register" message in every documents would be fine. I used SecuritySpy on the Mac, before deciding against it due to its appalling ergonomics.
 
Last edited:
I'm not in France :D I would bet French cheeses are specifically aimed at with such a legislation, as they are a "poor's food" there, compared to a luxury here.

Wow, another very long post!

Where are you then?

I meant you could buy the laptop abroad, trash the packaging, and just declare is as part of your luggage. But I read once tax law in Brazil was extremely complicated, in addition to unfair.

Yes, there is also this possibility. But it does not matter for the Brazilian revenue. If the customs officer finds a laptop in your baggage, he may request proof that either the laptop was bought in Brazil or that it was bough abroad and the Brazilian taxes were paid. I know people that were taxed at the customs office for a laptop they brought in another trip and others who were taxed for laptops they bough in Brazil but were not carrying the invoice with evidence of the payment of all relevant taxes.

The bottom line is, the Brazilian tax system is very unfair and complicated, and also very difficult to change (there is so much pressure from the government to raise the taxes even more instead of reducing them). Brazilian companies (and people as well) do not face the government as they should requesting taxes to be lowered. Not even almighty Apple, whose Steve Jobs once said Brazilian taxes were "crazy", chose to face Brazilian authorities. And Apple could befefit from that. Given the high prices in Brazil, the iPhone represents a very tiny fraction of all smartphones sold, regardless of being a very well desired product. And Macs represent about 1% of all computers around here, which is very low compared to elsewhere in the world. Should taxes lower, Apple could benefit of higher sales in a country with a population of nearly 200 million people.

I can't deny it, and referred to Ubuntu in its intention and core, but not its execution, while better than Windows XP, required too many fallback to the command line. I finally chose Mac because no Ubuntu desktop would be sold at a decent price compared to same laptops equipped with MS Windows ("The Microsoft tax penalty"), I refused to pay the Windows tax as it would come with Windows installed although I already had a separately-bought license (a hot topic on Linux forum is how to get refunded for that disguised "tax", which is unlawful according to many consumer's laws, but a de facto standard.), and that OS X had all the advantages of Linux, and added compatibility with commercial software should I need them. Top-rated service, software and hardware integration, and better overall engineering were bonuses.

I was skeptical of paying for use Windows. But after I tried Ubuntu and other Linux distributions, I found out how important software is. From then on, I pay happily Microsoft for Windows, Office and everything else I think is worth my money. I simply could not get along with Linux.

Agreed, MS Office is better on Windows, and I suspect Microsoft deliberately crippled the Mac version to push heavy users toward Windows.

I do not think so.

Microsoft Office is so good on Windows because Microsoft developed both, and was able to enhance Office so it could run seamlessly on Windows. In addition, it spends a huge sum of money every year on the development of Office (I read somewhere once that Microsoft spends nearly a billion dollars on Office alone per year, but I never found that source again). Anyway, Microsoft spends so much money on Office because it sells hundreds of millions of Office licenses per year and it wants to keep its market share high against the threats of Google, open source software and anything else that may appear.

Office for Mac, however, is a whole different story. It is made for the Mac. It was first made in Carbon, and then Apple transitioned to Cocoa. Developers had to migrate, and Microsoft Office is a very complex piece of software and, therefore, the transition should be hard. It should require a lot of resources to de-carbonize Office. In addition, Microsoft cannot afford to spend so much money on Office for Mac, as Office for Mac does not sell as much as the Windows version, and will never do.

Never had much success with OCR applications in Windows, though. A one-page typed text would always end up prompting 30-ish confirmations, with the suggestions wrong most of the time. One of my unsolved problems. I use Acrobat for that purpose on the Mac. What application have you used that yielded acceptable results?

I think ABBYY Fine Reader delivered good results. When have you tried OCR? I remember that in 1996 it was totally crap, but I was impressed with the results when I tested it again 15 years later. Did you find any good OCR software for Mac? I don't think Mac OCR software is on par with Windows OCR software, but I have not tested everything.

Not conditional, it is currently expensive because supply is low. But all things being equal, lower maintenance requirements on Macs would make the bill lower.

Yes, that is true, but you know this is totally hypothetical, right? The reality is much simpler: Windows is far more popular than Mac, so support is widely available for Windows and this trend is unlikely to change in the foreseable future.

Corporations and universities usually have a site license for MS Office. But given the current context of economic downtimes, these expensive site licenses have been put into question, as ditching Microsoft Office would result in millions in saved dollars. Our university has started to acknowledge the issue, and while they won't ditch MS Office anytime soon, at least LibreOffice is also installed so students aren't required to buy or pirate MS Office in order to be able to read and edit their files at home. I once wrote an extensive letter to the director of a national library he boasted aimed to help poorer classes access culture and computer tools. I argued that having only MS Windows + MS Office installed would not help disadvantaged masses to view computers as affordable, accessible machines, and that, by denying choice (a few Mac / Linux would have done the trick), denied democratization of computer knowledge, while normalizing piracy. It has to be understood that most used computers sellers here sell old machines at a high price, equipped entirely with pirated copies of Windows, MS Office being available on demand, all of it without proper keys. I think this is inherently wrong.

I understand that universities may direct their students towards LibreOffice or OpenOffice. It costs them nothing, and students can learn it and handle it. It would fit their needs.

Corporations, it's a whole different story. Corporations are there to make money. People on corporations know how to use Office. They know where they can find the features. Some of them were trained on that, and they may have years of practice. And the work is done faster because people know how to use Office. Plus, Microsoft offers support. Professional support. Suppose a company changes to LibreOffice. Yes, it may have all the features that one need, and it is free. However, there is a hidden cost here. People who have used Microsoft Office for the last 5 years may find it weird. They will have to learn how to use new software. It may be easy for you and me, but certainly not for an army of non-geeks that work at a company. Then, the company will have to provide training, which costs money. And that will consume time. Productive time. In addition, work may be done slower, as people are not yet familiar with LibreOffice as they were with Microsoft Office. If work is done slower, the company will have to hire more people or will become less productive. And there may be compatibility issues, and time spent to solve them. It all will add up to cost. The company may find out that buying Microsoft Office may end up being cheaper than installing LibreOffice for free.

There are initiatives here in Brazil aimed at promoting computers for lower classes. Some of these initiatives promote the use of open source software as well. Low-end machines running open source software could be the answer to provide computing power for the lower classes without having to lower the taxes that make high-end computers nearly impossible to afford. This strategy is, in my view, flawed. You can provide open source software for the lower classes, but then companies will still use Windows and Office. The people will end up installing pirated software which is widely sold in home-made CDs or DVDs. Or there may be some sort of social Apartheid. I don't think it will ever work. Sorry, I just completely lost my faith in open source software. I tried it, and got too disappointed. For me, it is made by amateurs for amateurs.

I have seen way too many people buying the small box containing Office 365 license having problems just to install it on current Windows 7 and 8, burning all three licenses for just one successful install to call it "worth it". Even the IT guy gave up on helping them since Microsoft obviously did something wrong, but provide no channel to contact them. Instead, he asks students to go back where they bought the box, and have it installed in store.

I bought Office 365 online and never had a problem installing it.

In another software category, I subscribe to SpiderOak, not because of its open-sourced-ness, but because it's tri-platform and secure. I mean, I happily paid for it at first, but since a year and a half, it can't re-download uploaded files; some do come fine, some others stall, for reasons their own engineers can't understand. They gave me a free year of storage, added 30% storage space for the same price, and they proposed that I take a whole new account, but I refused since they wouldn't be able to transplant the database on their end, or even mail me a physical disk with all my encrypted data on it. I don't have anything against paying for service, if only working! I can't even say an open-source model would have solved it, since I wanted to test ownCloud on my hosting space because it boasted on being a valid iCloud replacement with an option to encrypt files on the servers, but in reality, it repeatedly fails outside advertised boundaries, i.e. not requiring any root access to be installed and configured.

Yes, that may happen with some software. I guess, however, that is should not happen to widely used software such as Office.

We all agree on that. However there has been a long-time alternative to IExplore that is mature, flexible, and respects the user's privacy: Firefox. However it doesn't have any marketing strategy, and relies on word-of-mouth positive reviews.

I used Firefox a lot in the early days and I prefer it to Internet Explorer. However, it became such a memory hog on Windows that I decided to try Chrome. And I liked Chrome better. However, I do not see too many differences between them these days.

Let's be fair, it is claimed to do "80 to 90% of what MS Office does", the remaining 10 to 20% being mostly advanced functions we may or may not use. My giant spreadsheet was clearly in that exception range, but as very few pushed it so far, it was mostly unheard of, although way below the theoretical maximums allowed. I think there is a bug somewhere, because it can still easily crash on a machine with 16GiB RAM before saturating it. If I had to grow such a file again, I would probably choose Excel and add a CSV exporting step, or try to find a proper bidirectional ODS converter (to my knowledge, Gnumeric has its own format (!), and supports ODS as well). Added manual work, but probably better efficiency overall. However, given MS Office nasty tendency of modifying files and formatting automatically against a user's will, I wouldn't trust it too much. For that reason, while I would go with Excel for performance reasons, I would stay with Writer for the thesis; neither it nor Word are very good integrating images in the text, and perhaps only Pages does it right, at the expense of non-existent reference managers support.

Actually, it is hard to tell it has 80% or 90% of the Office functions because I have not yet seem all the functions mapped.

Still, there are a few useful features that LibreOffice lacks. For instance, Word's draft view is the best for writing in my opinion (especially with footnotes), and LibreOffice Writer lacks it. LibreOffice Writer also does not allow to split windows so I can see and edit two parts of the same document simultaneously. This is a very useful feature in Word.

This is probably my biggest gripe against MS Office: too many bugs disguised as functionality, and just like Windows, settings are difficult to find, not mentioning non-existent support.

Never used Office support. And I guess Office has much less bugs now than it had ten years ago.

If a corporation wants me to use MS Office, it will have to provide it. Otherwise this amounts to pay for the right to work. LibreOffice isn't a "cheap" alternative, it is a full-fledged office suite that may or may not meet one's needs, free of charge, and with a large community support. Not to say I praise it (in fact I am still waiting for an answer on proper ownCloud alternatives – doesn't behave as described: "you're doing it wrong" was I told –, and a configuration guide for OpenVPN), but at least in this case it's efficient. Note that also applies to any job requiring the applicant to have a specific tool, without compensating properly for the added expense, e.g. common "delivery man with car needed", and no mention of a per-km paid expense.

Yes, LibreOffice is a full-featured office suite, but Microsoft Office is widely used in the enterprise. I like LibreOffice, and it is a good piece of software, and it is surprising that it is free. However, I wish the pace of innovation was faster. Still, I feel Microsoft Office is better. It is quite unfortunate that LibreOffice has to compete against the biggest-budgeted software ever.

One would have to find it first. Default MS Office settings, just like Windows, can be customized to infinity, and no installation is identical to the next one, especially in a large installed base of Microsoft technologies. This probably contributes to the high manpower required to maintain such networks.

Oh, come on. Have you ever tried to open an OOXML file with WinZIP or a similar program? It is a ZIP file, much like ODT files are ZIP files too. In both OOXML and ODT, the file contains two files inside it: one with the text alone and other one with the configuration and page layout. You can open the text file with any text editor. That assures that, no matter what happens to Word or LibreOffice, you will always be able to recover the text.

Sure. But the same could be said of software companies that completely dominate the market: whatever their users are thinking, it's not like they are going to go anywhere soon since they would still pay for the software in the form of bundled sales, and that hardware manufacturers aren't free since cutting ties with Microsoft would result in higher prices due to M$ not paying part of their marketing expenses. That leads me to say most people don't "choose" Windows. They rather "default" to Microsoft buying a price that suits them, without consideration for future maintenance costs. I strongly advocate that the laws forbidding bundled sales should be enforced, and that a given computer be presented with a choice of OSes, separately listing the license price for proprietary software, unless the same manufacturer did both hardware and software. Sure many would still choose Windows, but at least choice would be provided, much like the legendary Windows version that asked users which browser they wanted to install first. I never came across this version, though. At least there would be competition.

You appear to be very anti-Microsoft.

Microsoft does try to please its users. Otherwise, it would be doomed. Microsoft has to sell a lot of copies of Windows. Windows may come installed in new machines, but people are buying less and less computers. Microsoft makes a lot of money selling Windows to users who already own a computer. Windows 7 was a massive success, but Windows 8 flopped. Windows 8 still does not have a significant market share, and that means that Microsoft is earning less money. For this reason, Microsoft fired the guy behind Windows 8 and is trying to fix it.

Microsoft also has to enhance Office so people keep upgrading it. If a new Office suite is not compelling to the users, then they are not upgrading it and Microsoft is not making money.

So, software companies, even those who dominate the market, try to please customers. They want people to buy software over and over again, and the next product is always competing with the last version.
 
A minority of open-source developers do care about the user experience, and mostly develop on OS X. Oh well.
Well as I said I haven't got anything against paid-for software. RedHat is in fact a company that makes money providing premium support for its software, while leaving the source open for others to use, calling it Fedora. Ubuntu was criticized by some purists because it originates from a for-profit company, Canonical, that imposed its design decisions (M. Shuttleworth once declared that "every software that requires the use of a command-line on Ubuntu is a bug that needs to be tackled". Some were outraged, I was in bliss). I feel this is simply stupid, as everything is still open. Even the Linux kernel isn't democratically-developped.

I do have, however, a fear of locking myself out of my own files because the format wouldn't be interoperable, and the software editor may choose to cut support for older versions, even remotely. The same reason was invoked when the EU wanted to pass into law that all government-issued files had to be available in open formats. This was also an argument in the StopPalladium campaign (early 2000s) against the the TPM to be integrated in future hardware, with the rumor that Microsoft would use it in its Windows to arbitrarily lock access to a user's own files. In the end, this opaque chip is now integrated in most computers, but editors so far haven't made the jump to use its full capabilities.

Note this isn't specific to Microsoft: Rosetta, as a piece of proprietary software, was never released as open source for developers to build upon since its license allegedly forbade it. The few PowerPC proprietary software remaining are the other half of the problem, as releasing the source would at least allow for continued functionality. Let's not forget that Darwin, OS X's core, is also open-source. Maybe part-time developers haven't build anything on top of it because OS X was so good from the start :D As you point, this may be the right road to follow:

I am not anti-open source or anti-free software. I am anti-ideologists who claim that Linux is better than Windows just because Microsoft makes Windows and Linux is free and open source. Some of them voice their opinion against those bastard capitalists who want to make money out of software, and complain of people who don't see the truth behind the beautiful open source software which is a thousand times better. The same can be said about them bragging about LibreOffice against Microsoft Office. Technical arguments are set aside, but they still praise open source software as technically superior, even though it is inferior in several aspects. I just hate it.

In fact, given all these examples, I'd rather have development of both FOSS and proprietary software revolve around the use of common, open formats: ODF is criticized for allowing to many unspecified elements (probably why Gnumeric can use it, and Calc just not interpreting these elements), and OOXML to be intentionally bloated to prevent anyone else than Microsoft to implement it properly.

I've heard that.

:confused:

AFAIK you would have to redo all your bibliography again since EndNote refuses to use open standards for storage. That is one tough choice I had to make when choosing a reference manager: while I was sure I didn't want EndNote (chiefly because there was no LibreOffice compatibility), I wasn't so sure about the others since, while open-source, they wouldn't necessarily use the same formats, nor interact with one another. So I more or less defaulted to Zotero as it was available on all platforms, and that Mendeley, while more ergonomic, lacked compatibility with LibreOffice.

As for bibliographic managers, I guess you have to choose one and go with it until the end. I found Zotero to be nicer than Endnote. Endnote's engine seems to be ancient. Zotero seems a better solution, but I found its styles (CSL) very difficult to edit.

Antidote is an example of such a software. Proprietary, but respecting users' choice, although support isn't that good.
Well I wouldn't pay premium price for a software I am not even sure would be useful, as I'm not a writer. It wouldn't be wise. But a fully-featured version that would include a "Please register" message in every documents would be fine. I used SecuritySpy on the Mac, before deciding against it due to its appalling ergonomics.

I bought it with a discount, otherwise I wouldn't have. I usually buy software I don't need when they are sold in bundles that make them a bargain.

PS: I had to split this post in two because it was so big...
 
When did Sente go free?

Sente is not free. Well, it is free for libraries of up to 100 references but, if you need anything above that, you must upgrade to a premium account. I don't know exactly when that happened, but it was a while ago.
 
Never even heard of DevonThink Pro before this thread. But won't be trying it either as it costs money - Mendeley is free and does all I need!

Comparing Mendeley with Devonthink does not make any sense at all. Devonthink is not a reference manager. Not your fault for making the mistake of comparing the two. That being said, Devonthink is an incredible piece of software and I encourage everyone to give it a try.
 
Comparing Mendeley with Devonthink does not make any sense at all. Devonthink is not a reference manager. Not your fault for making the mistake of comparing the two. That being said, Devonthink is an incredible piece of software and I encourage everyone to give it a try.

Yes, it is true. Said that, there are not too many programs similar to DevonThink. Perhaps you could compare it to Evernote or Yojimbo or Eagle Filer to Together. Mendeley is also a very particular piece of software, although its bibliographic management capabilities make it more akin to Endnote, Zotero, Sente, Bookends, Papers and others.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.