Despite the growing success of the Open Source movement, most of the general public continues to feel that Open Source software is inaccessible to them. This paper discusses five fundamental problems with the current Open Source software development trend, explores why these issues are holding the movement back, and offers solutions that might help overcome these problems.
Fundamental Issues with Open Source Software Development
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
I had written a big todo about this. But heh, if he really can’t work with most the interfaces….. I am guessing the project he is speaking of must be particularly awful.
They are right though, it’s their code and they can write what they want to. If ya don’t like it….well don’t use it. There’s a lot of things I don’t like, I don’t use them. And if I have to, I learn to like it.
Actually, the author is a woman. I found the article actually had a lot of merit – it could have been better by including real-world examples for each point, though I understand why she chose not to name the actual program she took as a primary example.
There ARE user-friendly open-source apps out there – the more we reward the developers by using them (and telling them about it), the more the trend is likely to catch on…
i always thought open source was popular because it was free. You usually can purchase software with open sources (or called shared source) that lets u modify it and use it within your own enterprise but not distributing it.
I find it very tiresome to read the aged usability rant again and again. This might have been true a while ago but people talked about it, people fixed it. If you look at any modern open source project, I’m sure you will see a lot of focus on good usability. The author mentioned Firefox as a positive exception but heck, Firefox has one of the _worst_ interfaces of an open source application which I’m using today (but it’s still really good).
When I compare the usability of my average new GNOME application to the quality of the average closed source application (which I used) then no, the open source applications don’t lose out. Granted, I could be blinded by zealoutry and I can’t proof the opposite, but nobody can deny that there hasn’t been at least an improvement over the last years! Anyone who doesn’t even mention this “usability movement”, doesn’t seem to have a lot of insight in my eyes.
It is ok to mention it again and again because it’s really important and some are still doing it horribly wrong (like Project X, obviously), but it’s not ok to completely disregard any of the improvements which have happened during the last two years, not just in application usability but especially in usability awareness among the developers.
It’s nice to see some people know how to write a real paper still. Not just a rant with no real support or proper formating. No attacking, just a solidly written paper.
And everything the person said is right on.
Reasonable, well articulated critiques of open source software are rare birds indeed.
Kudos to Michelle.
Nicely written article. I agree with most of what she had to say concerning OSS. Some OSS developers make an effort to include detailed docs such as for the KDE desktop or apps like GIMP. Then there are others that look like they were developed for the tech in mind with out including any help docs. To actually compete with other platforms Linux first has to actually have a standard set. Developers should also learn from the competition and try to improve on their mistakes not create new road blocks for users to over come. I’m not a programmer but it seems simple enough to understand they should design software for the intended market. If it’s based on a Windows app then it should have the same feel and features of the Windows app, not less or more complicated GUI. Just offering it for free is not always going to attract others from a working app they already have.
compare this:
“…we were carefully instructed that the best software tools are small programs that do one thing well and interface cleanly with the other tools.”
with this:
“..the [Linux] developers want to create this massive uberframework that connects to everything, is loaded with features and [presumably] out performs Windows…”
and it certainly verifies Michelle’s thesis.
That’s my complaint exactly. There is no real world support, we are just supposed to nod and say yes by plugging in our own disliked project in place of project “x”.
I’ve ran into bad documented projects before, but I’d love to know what her project “x” is. She obviously misses the point of being open is that when your project sucks…people say so.
…explores why these issues are holding the movement back…
I really dispise this phrase. Who says that the movement is being held back? Most of the world can’t operate a Sparc system or install Netware or configure Windows 2003 Server, but we don’t blather on about those projects being held back (no offense to the author of the summary, its a commonly held sentiment that in my opinion has no merit).
…because she hit the nail on the head!
“Who says that the movement is being held back?”She is evaluating a desktop Linux application. I was under the (wrong?) impression that most people are aware that Linux only holds a small percentage of desktop marketshare. Even though last year and the year before were the year of the Linux desktop.
It’s true what you say about Sparc, Netware, and Server 2003, but then again, these are highly specialized systems not meant for the general public. Most OSS IS out there for all of us to use. Yes, some of it is quite simple, but some of it isn’t. Much that isn’t is painfully bad. Many of the features I use often were so counterintuitively buried in word processors such as AbiWord, Oo.org, and the old StarOffice that it took me a lot of time and frustration to figure them out. Notice I say “counterintuitive” and not “different:” I understand that there’s more than one way to skin a cat…I’ve used both Word and WordPerfect for many years, since long before they had extremely similar feature layouts. But when it takes me *half an hour* to figure out how to number pages and leave the number off the first page, something’s terribly wrong. And THAT is what is, in fact, “holding [many of these apps] back.”
The complaint that the Linux desktop is not well supported is something that can be fixed. The OSDL is only starting to influence what makes a Linux desktop.
The arguments are valid but it is like an ecosystem, the fittest will survive. But the abbundance of lifeforms makes for really fit applications. Time is on the side of evolution.
GerardM
“Hey, I saw a black sheep yesterday! Thus most sheep must be black, too.”
It looks like the usability paragraph (and some of the others) have been inspired by a conclusion like this.
Additionally, she seems to have used Debian Stable for her test. Stating GNOME as a “clunky desktop” might have been true for the 1.x series IMHO, but it doesn’t hold for the 2.x series anymore.
However, the most annoying thing is the underlying assumption that the Open Source movement has no constraints on resources (or at least as much ressources as the “usual” software development model). This is simply no true.
The better the financial (corporate) support of projects, the less of her points are valid. And it makes sense that small, less supported projects have a problem reaching the same level of professionality on areas not directly related to the code of a program as supported (or proprietary) ones.
Thus, rasing awareness of these issues is a good motive but it misses the central problem, IMHO.
One thing that took me a long time to figure out in OpenOffice.org/StarOffice was how to insert a page break. It’s easy in Office, you just click on the insert menu, click on break (the first entry, btw) and hit ok (since page is the default selection for the break). I haven’t used OO.org in a while so I don’t know off the top of my head how to do it. I personally prefer Wordperfect (really wish I could get a hold of the Linux version, even though it’s probably outdated by now), though I don’t do a lot of typing. The other day I was typing a paper and wanted to insert a page break and for the life of me, I couldn’t figure it out in WP either. I searched through the Help to no avail and unfortunately my friend, the WP master, was offline. I ended up just hitting enter till I got to the next page.
Simple things like this are what really bug me sometimes.
One time I spent around an hour trying to figure out how to turn off transparency for menu’s in KDE. I went through all the options many times in the Control Center. I just didn’t see it, though I knew it was there since I remember going through the same thing trying to figure out how to turn it on after re-installing.
Theses examples tend to be the exception to the rule though and I don’t have many usability issues with any software.
My mother has never needed to work with a computer until recently. She was lost while sitting behind the screen for the very first time as any other person in the same situation. Didn’t know what to press or do. No matter what the OS is.
It is a question of practice and getting used to than layout or EVEN intuition. It is a question of those first few courses that help you memorize the basics. Once you have learned it and computers are not your hobby absolutely. You glad you know those few steps to achieve your goal. Then you don’t even want to transfer to another system because that would of course mean starting from scratch in many aspects….
I use Fedora with Gnome and it is organized, intuitive and with OpenOffice I do not need MS products to work. AND ALL IS ABSOLUTELY FREE!! There’s also support, but let me stop now for that might be a long story.
That was a good article. But – and though this is a bit out of topic – why does OSNews want to publish so many news about the supposed problems of open source development model etc.? (There have been some links to articles that have had more FUD than relevant content, you know.) And why there seems to be so few news and articles about the problems of proprietary sofware, or problems related to sofware patents, and such issues?
Has anybody else noticed this kind of trend in the OSNews news policy?
For example, nowadays strong forces in the EU are proposing strong software patent laws that is a serious threat not only to open source but software development in general, worldwide. If this dangerous directive is adopted without proper amending, most software will become illegal to use in Europe. The same elsewhere, of course.
Many open source related websites are now protesting against software patents in Europe, and have taken part in a web strike. See for example the Debian homepage: http://www.debian.org/ I’m sure that it would be too much to ask OSNews to follow suit in that sort of a protest. But at least OSNews could, for example, publish sometime some news and articles about those kind of really serious threats and problems open source is facing today in the world. If not, why not?
FFII: Software Patents in Europe:
http://swpat.ffii.org/index.en.html
What can we do?
http://swpat.ffii.org/index.en.html#gunka
How you can help us stop Software Patents:
http://swpat.ffii.org/group/todo/index.en.html
Specially Recommended Reading:
http://swpat.ffii.org/index.en.html#links
OSNews:
As far as I can remember, there have been practically no news about this kind of subjects, the really big problems that open source operating systems like Linux and *BSD are facing nowadays. Why?
As far as I can remember, there have been practically no news about this kind of subjects, the really big problems that open source operating systems like Linux and *BSD are facing nowadays. Why?
Well, perhaps its because no-one has written them, or maybe you missed certain discussions on software patents.
Like anything, osnews is a community. If you feel there should be an article regarding these issues, as outlined in the links you provide, perhaps you should write one?
Its always seemed to me that Eugiena and Co. have had a chip on thier shoulder for OpenSource software. Besides, OSNews is a business, and controversial articles get more viewers.
As for the article:
If I’d put the same person on KDE or Gnome, they probably would have spent half of their time fighting their own intuition…
I really find this statement bogus. Is she saying, that just because users are used to the way something works (intuition based on past experience with Win or Mac), that is the correct way, and it must be followed, always? If that was the case, we never would have gotten a GUI (the CLI is what everyone’s used uptil now. That must be correct).
Or, is she saying that people have some sort of natural intuition when it comes to funcitioning with computers? Sorry, I don’t think so. Just about everything with a computer is a learned skill. A previous poster mentioned he had trouble finding a Line Break in OO.o. We’ll, it maybe be intuitive to him to have that entry in the Insert menu. However, it could be intuitive to me for it to be in the Edit menu. Who’s right?
Now, I’m not saying that OSS usibiility is perfect. It still has a long way to go (but, its at least on par with most closed source apps). I do, however, feel that docuemntation should be unnecessary. If the usibility is good enough, everyone should be able to figure it out on their own. Besides, most end users don’t read documentaiton anyway. Only us geeks, who are determined to get something to work will read documentaition.
I’m very surprised to find that some people have praised this paper not only for its content, but for its merits as a paper. I have to strongly disagree! Many of Michelle’s statements are un-supported, and few real world examples are given.
Before I go on I’d like to draw attention to Michelle’s footnote:
“In this paper, I use the general term Open Source, though often I’m exclusively discussing Free Software. As well, when I use the term Open Source projects, I’m usually referring to projects that have a contribution base wider than one or two individuals. I’m also aware that some companies release Open Source versions of their software, and though I certainly appreciate their donation, I’m excluding these Open Source projects in this particular paper’s definition of Open Source, as some of my statements do not apply to them. I made these generalizations for the point of simplification, and not for any political motivations.”
Make of this what you wish.
I’ll make a start.
Introduction
“most of the general public continues to feel that Open Source software is inaccessible to them”
What evidence does she have or present? Most of the public I know are unaware of the existence of Open Source Software”.
“…explores why these issues are holding the movement back, and offers solutions that might help overcome these problems”
Holding the movement back in which way? A new version of the Gnome desktop is released around every 6 months, making great improvements every time. Kernel development is continuous on many fronts and at what seems to be a good pace. If she is referring to market share, then in my opinion the reasons she give are in-significant compared to many other issues.
User interface design
“I’ve seen people with relatively little computer experience navigate around the OS X desktop for a few minutes, and then turn around and tell me that it “flows very nicely” and “just feels nicer” than what they’re used to. If I’d put the same person on KDE or Gnome, they probably would have spent half of their time fighting their own intuition”
So people have told her that OSX is intuitive, whilst she imagines that the same people would not be so complementary about KDE or Gnome – guess work and un-supported opinion. In my experience, a modern Gnome Desktop is intuitive to existing Windows users. For example, my partner has been using Gnome, OpenOffice, and Ximian Evolution for around a year now, and has received very little assistance from myself (virtually none). At work she uses Windows 98, MS Office and Outlook. When questioned a few days ago about her experiences she commented on how easy she found it to use the software, and that her Windows learnt skills were easily transferable.
In my opinion, as an experienced professional IT trainer, Gnome applications are more user friendly than Windows equivalents. OSX is very good, and the design is to be commended, however I don’t think that Windows learnt skills are as easily transferable to OSX as they are to Gnome. I’ve experienced this with two of my non-techie friends who went Mac and found the transition difficult.
I won’t comment on general Open Source user interfaces as I’d need to evaluate many applications, but for the ones I use which are mainly Gnome Applications I think they’re excellent.
Documentation
“Open Source projects tend to have a major problem with providing decent documentation”
Which ones? How?
Many OSS projects have good documentation, Samba documentation is meant to be good for example (as stand alone chapters). Man pages are sometimes excellent, sometimes not, Debian ones are generally excellent.
How many people actually read documentation anyway? It’s often ignored, particularly by men. It’s interesting that this article was written by a woman! Seriously, Michelle’s general gripe about documentation is not specific enough to be helpful, does she mean user documentation like help files, admin or developer documentation? Examples are few again.
“they’re found in Usenet articles, bulletin boards and chat logs. Users who can’t figure out how to do something runs to alt.projectx.devel and asks the same question as hundreds of users before them. Some expert takes pity on their plight and responds to their question, but never documents this answer. So when the next user hits the same problem, the process has to be repeated”
Forums are searchable so questions and answers are seldom repeated. Forums are quick and usually helpful, these are the backbone of the Open Source movement – in my opinion. Michele seems to have some misconceptions, is portraying a different worlds to the one I know, or has different experiences – if this is the case she doesn’t share.
What’s the Microsoft documentation like anyway? What is their support like? OSS has much more helpful support through forums etc. than proprietary software does.
Feature–centric development
I’m sure that some projects are as she describes, though some of the scenarios seem to be of small OSS projects that are installed from source (as opposed to a component of Gnome installed via synaptic and apt-get for example).
Programming for the self
I’m not really sure what she’s on about here. Yes programmers in the OSS world create things for themselves, but many are interested in the users, take the Gnome HIG for example.
“The result is that Open Source projects are made by programmers for programmers”
I’m not a programmer and I don’t have a problem. My partner certainly isn’t a programmer and she doesn’t have a problem. Michelle, please explain what you’re on about!
Religious blindness
“While this has the advantage of increasing Open Source software usage amongst programmers themselves, unfortunately it has the side effect of preventing the Open Source community from learning what proprietary software has to teach”
This may have been true once, but certainly isn’t true anymore. To use Gnome as an example once more, they have taken what they consider to be the best ideas other Desktop’s Propriety and OSS. It’s certainly not just programmers using OSS these days either.
My Conclusions
Michelle doesn’t raise or fails to raise many useful problems and solutions in OSS today. I’ll give my 2p worth when I have the time.
One of the main reasons I prefer not to use Linux and open source applications for it – is the lack of binaries. The developers feel (reasonably) that they cannot provide an executable for each and every Linux distribution; so what you are offered to download is the source files and a make file. You yourself have to compile and build the application if you ever want to use it – something I’m sure my Mom and even more experienced average users won’t go through. It what holds the programs from being used in the real world. What user wants to learn the structure of C-based applications, the ins&outs of ‘make’ and ‘.configure’ and then to cope with errors thrown by the compiler – while all he/she wants is a working application?!
Think of it in “physical environment” concepts. What if any furniture you buy you must assemble yourself? You order a refreigerator, have to come with your own car to take it out of the shop, then at home you take all of the parts including separated bolts and must know how to put each of them at their place. Including having the neccessary tools to do it.
Sorry, but things do not work this way for the overwhelming procentage of the people out there.
I’m frustrated with this process of ‘make’-ing every program and I know I’m not alone. The Open Source community can be deaf to these voices – but, well, that’s your problem. Same as the fact you cannot provide a binary file for every Linux distribution – you created this fragmentation of linuxes so deal with this problem. Why should the end-user be the victim?
If this community wants to ever enlarge the usage procentage of products being used in the real world it should start being much more user-centric. If only money what makes you do it – then I’m on the side of proprietary software.
People are talking about (at least) two different “movements” when they mention GNU/Linux. One movement is the traditional software should be free-I am doing this because I want to approach. The other is trying to make money.
If you are an independent who really doesn’t care about making a profit, then it makes perfect sense to design for yourself and/or your buddies. There is no reason not to. Certainly if you are not getting any reward, tangible or otherwise, for making the extra effort then why bother?
The gripe I have relates to “commercial” OSS, specifically Linux distros and apps, that really don’t offer any value for the money. Consider please, why would anyone pay money for something that they can get for free? Answer, convenience and support.
Usability is only one facet of the overall deficiency in customer service and customer relations that the commercial OSS industry suffers from.
If you pay cash money for an OS, especially if you are coming from Windows or Mac, then most of the users I personally know will expect a reasonably polished, effectively integrated, and _properly_supported_ set of software.
If someone pays anywhere from $25-$100, or even more, for a commercially packaged box of software they expect a little more than for their money than, “If you don’t like it, don’t use it fool. RTFM.”
Especially when so many commercially Linux distros lately keep claiming that they are as good as windows, as easy to use as windows, as intuitive as windows, as pretty as windows, but a whole lot cheaper. What is a newbie supposed to think when they see that kind of false advertising?
They will think that they have been screwed when they take it home and it won’t install, or it won’t run, or it doesn’t do what they are led to expect. Then, when they ask the company that _they_paid_money_to_ for help and get a casual brush off and/or get advised to go talk to other users, it doesn’t give a good impression. Not a good impression at all.
Some commercial distros even limit their support to email only, then put a limit on the number of emails. This is insulting to a paying customer. Not to speak of short sighted. One pissed off customer can cost you a dozen more. How much effort is it to read an email and reply? Seriously?
Free OSS is exactly that, free. So you take what you get and shut up. If you have a problem then write a polite request for help and you might get some. But no one has any “right” to expect extra effort on the part of a developer who gives away their work for free.
HOWEVER, if you are going to advertise your OSS as being commercially viable, if you are going to make publicly documented claims about the quality of your commercial offering, then you better be ready to back up those claims. Unless you LIKE the idea of financial ruin and bankruptcy. That means taking the time to develop software that a non-geek can actually use, and then supporting it.
Free software developers don’t owe anyone anything. We all owe THEM for the gift they have given us. But commercial developers DO owe their customers some consideration and some hand holding. If they want to stay in business.
Lyn, I award you five stars. Michelle’s paper is neither analytical nor critical. It is an eloquently camouflaged rant with little to no facts, illustrations, or even credible references.
She makes broad generalizations, gives vague inductive premises and provides no example of why “open source projects” are, in her words, “holding” themselves “back.” Her paper would have held a little water if she meticulously scrutinized, at least, one popular open source project. But alas, she failed to provide the most basic concrete evidence that would have lent some form of validity to her paper.
In a critical thinking class, the paper will score a grade of F. Her paper is equivalent to the shallow comments one witnesses on osnews. “All rants, little substance.”
I love it… and I love OSS
anyway, this does my baps in…
article writers do a review on a piece of oss, now, most of these “computer experts” have had a few years of working in an office which runs Windows pcs. These people “know about” computers and computer software etc.
Now, bearing in mind they come from a Windows background, they think the bigger the version number, the better the software will be, so they download the latest version of an app… the app plays up erratically and the writer puts out a damning review of this buggy software.
NEWS for you Mr Windows User
99% of the time, bigger numbers are DEVELOPMENT versions.
e.g.
my_excellent_app.i386.1-3.0.4 (developer version)
my_excellent_app.i386.1-2.1.7 ( STABLE version)
note, the STABLE version should be downloaded for day to day use, and the developer version should be downloaded ONLY to help the author(s) in finding bugs etc.
Hope this helps clear up some things for some people.
“What if any furniture you buy you must assemble yourself?”
have you ever heard of IKEA? the owner of IKEA is the richest man in the world actually, it seems like this “do it by yourself” method works also for forniture stuff…
and anyway, you don’t need to use the source, if you want there is debian out there, so many binary packets that you’ll never need something else…oh and if you stil want to use the source, use gentoo, emerge <your fav app here> isn’t difficault at all…
please, ppl start using your mind and don’t think the user are so dumb they can’t do anything…
A lot of these complaints apply to commercial software
too. Why single out the OSS?
“please, ppl start using your mind and don’t think the user are so dumb they can’t do anything…”
And yet that’s the foundation for many anti-linux rants.
Quite frankly, if I was an “average user”? I would be insulted by the characterizations.
@Lyn Rees
FOSS documentation is often too geeky. Or you have to search and google for pieces of it here and there, and possibly still do lots of distro specific experimentation etc.
But actually the real problem may often not be so much the need for documention (there is that need naturally though), but that the FOSS user interfaces are often too difficult to use without first reading documentation. So, just like Michelle says.
An example (there could surely be hundreds more):
How do you properly configure SpamAssassin without reading its documentation? No way. But the documentation of SA would be all too difficult to understand for your Aunt Tillie, don’t you think? Geeks and nerds usually do just fine with things as they are (like with SA), but not Aunt Tillie, or even his more computer-savy son Joe Sixpack…
So, better documentation and better user interfaces are a real need in FOSS.
GNOME and KDE are quite good nowadays though, in both respects: user interfaces and documentation. But usually not everything that Aunt Tillie or Joe Sixpack needs can be found in KDE or in GNOME.
How do you configure the necessary firewall, for example? Or install and update certain extra software? Make backups of your data? Burn A DVD? Watch a DVD movie on Linux?
Aunt Tillie or Joe Sixpack cannot solve such problems easily, especially if there’s no easy enough documentation available, and not enough people interested in writing such documentation.
This was a good article. I’m sure we’ve read any number of articles talking about open source documentation/GUI/intolerance… so that’s all in the past.
The one very good point I think this article raised is the issue of bloat/feature creep. I know, I know…this should be reserved for critism of windows
But it is in a sense very true. I use Firefox almost exclusively now. Nice program, but look at all the ‘other programs’ embedded inside it.
Its got its own HTML source viewer, download manager, Calender applications, games, calculator. Now I know only the first 2 come in the default install. But when the solution to a lot of requests is: GET SUPERDUPER PLUGIN this has the same effect as feature creep. Its the culture surrounding this that often violates the principle stated by the author (Do one thing really well).
Does everything really need to be its own mini operating system like emacs?
Plugins are a nice idea, but when they’re so much in the forefront of the user, it can have detrimental effects. It seems interoperating easily between applications (a la copy/paste, DDE, networking, document objects…) has been somewhat lost.
“One thing that took me a long time to figure out in OpenOffice.org/StarOffice was how to insert a page break.”
…you’ve got to be kidding. In OpenOffice, click Insert->Manual Break. Heck, it has the word “break” in it and even is the first menu item!
And just because you can’t figure out how to use something the first time you look at it doesn’t mean it’s usability is bad. No total newbie computer user can operate either Windows or MacOS X.
I disagree. I firmly believe that the “holding back” that people are referring to is synonymous with “isn’t Windows”. Since that is the case, Linux will always be “held back” because Windows isn’t something Linux is trying to be (otherwise Linus would have called his OS Lindows).
I’ve used Linux as my primary desktop for about eight years, and I have never once felt held back by the OS or any of its tools. I’ve used it as a server for longer than that, and have never been “held back”; quite the contrary, which is why I continue to use it.
The fact of the matter is that Linux is a highly specialized system that has a lot of power and features. I guess I just get tired of people gearing everything to the dumbest common denominator.
“Intuitive” is perhaps the most overused word I’ve ever seen used yet on this forum. It’s so overused that the meaning of the word has become lost on the lot of you.
What is intuitive to one person will not be intuitive to all people and possibly not even to the next. To believe that people can use any device without knowing even the slightest amount about said device is ridiculous. Could you use a shovel without knowing what it was for before hand?
Can you drive a car without having some clue to it’s operation? Is it “intuitive” to have to press the accelerator different amounts at different times to increase and decrease speed? Well, that depends on who you ask of course; however, nearly everyone (where I’m from at least) does eventually learn to drive. While not everyone becomes as proficient at it as others, they can get the hang of it.
Not everything can model physical objects that we can apply our “natural tendencies” (whatever those are) to. If we only relied on concepts we could describe in that way, where would be? Math isn’t based on intuitive concepts which is exactly what allows it to be so expressive. That’s its strength not it’s weakness. Lets not forget that another definition of intuitive is “something that you don’t have to think rationally about”.
Technical people aren’t going to want to use the same system that non-technical people are going to want to use. Why should they? Why not just use document formats that are independent of the program that created them? I don’t care what program you use to publish information, just don’t send me a word document and expect me to read it.
Now, let us not take this out of context. Software can always be made to be better, and even easier to use if the situation allows for it. Just stop talking about intuition as if it has any real rational meaning.
I knew while I was reading the article that someone was going to post a comment to the extent of “If you don’t like it, then don’t use it!” And sure enough, there it was, the very first one. I guess that brings us to the real issue of what’s wrong with the open-source community. There are a lot of programmers who contribute to, or even worse, manage software projects yet have the mental maturity of 12 year olds. That pretty much explains everything she critiqued in her paper. Smart guys, bad manners. Plain and simple. And of course that doesn’t apply to everyone, but unfortunately there are enough of them that it gives free software a bad after taste.
The only thing I disagree with the author and a lot of other free software advocates is the idea that Aunt Tillie should be able to use Linux. I don’t see this as a good goal to have. Aunt Tillie already has an OS for her, it’s called Windows and she can buy it for 200 bucks. Though I do think that progress needs to be made in regards to user-friendliness, it shouldn’t be in the direction of having every moron be able to use this software. I think that the target audience doesn’t necessarily have to be programmers either, just people that don’t mind having to do some thinking. There are a lot of technically adept computer users who don’t know how to write code but find themselves frustrated and lost when trying to switch over to free software.
“Make a program that any fool can use and only a fool will want to use it.”
How about the simple fact that you can’t link against anything under the GPL unless you want your code too to be GPL? That may be fine in RMS ideal land, but in the real world where each business has unique requirements for its market, this is often an impediement. How so? It means that if you want to realistically incorperate a GPL component into your code (and there are some great GPL components out there) you have to either fork to it as a separate task, or make your whole app GPL. This causes a lot of “re-inventing the wheel” in software land, to get around this issue, which of course goes directly against the desire to re-use as much code as possible. Worse, many young developers of Open Source software pick the GPL by default because they don’t know any better, and then when they try to change terms of their liscense they face a revolt from many in the OSS community.
The fact of the matter is, Open Source software has many merits, but to expect large software systems to never interact with close source is unrealistic. This is a large impediment that will prevent many commericial ISVs from using OSS. And right now OSS could use some large commerical ISVs adopting their technology.
That’s why the LGPL exists. You will see some GPL libraries by people who want to support only Open Source projects, but there are a lot more closed source libraries which aren’t free to use for everyone either. That argument is really bad.
Every core library on a modern OSS system is LGPL or less protected (if you take aside Qt, which is also available under a less protective license if you pay for it).
I agree, intuition is greatly overused in GUI discussions.
If intuition was so important, how did the video cassette recorder ever make it into the market? No two brands look, work, or even behave the same way. Yet everyone manages to program them to record a movie.
I routinely use three different OSes (Windows, SOLARIS/Linux – don’t tell Sun!, and a weird proprietary thing I design in). It was not fun getting the hang of them, but now I don’t even think about switching between one or the other (and I still occassionally use DOS too!).
Intuition is really nothing more than a fuzzy memory that “points” you the right way.
Cheeers
Yohn
Thanks B.Smith for saving me the time of writing what you have already written.
There is nothing else to be said about this matter. It is insane to think that a developer who is doing something and giving it away for free to who ever might find it useful should not do what ever he/she pleases. The person has the personal freedom to to make the user interface a square of black with black type over it and the person taking this program for free has absolutely zero right to complain.
If I give someone with bare feet a pair of shoes and they complain am I expected to find a different pare of shoe to give that will make that person happy. No, I would snatch the shoes back and give them to whoever will find them useful. Would Michelle Levesque really expect me to have an obligation to cater to who is taking for free?
Who ever said that OSS wants to achieve any of the goals that are assumed as true in the article. The individual developer makes the decession about what the goals are for the piece of software he/she is developing.
I also completely agree with the idea that when it is being sold the seller is then held accountable for the types of things mentioned in the article. These are seperate ideas.