Propose your solution
Attachments
No attachments.
Duplicates
Comments
|
|
I'm laughing that someone raised bug #1 on Launchpad!
Solution #2: As far as I know, it is Microsoft, not tax-supported educational institutions, which offers the discount in order to get students used to its software. You can see this on its website where it sells its products at a discount to students.
Therefore, I would suggest that raising a petition would backfire.
|
|
|
|
@PaddyLandau: I'm not sure I understand what you mean. If I understand correctly, Microsoft offers their software at a discount to the educational institutions (including students), which means that even the academic organizations themselves are able to buy the software at a discount (but still have to pay for it). So, even though it's cheaper, it's still taxpayers' money being used to purchase these software licenses. But please correct me if I'm wrong.
|
|
Auzy
wrote on the 28 Dec 10 at 05:59
|
|
|
Some unis are already running Linux as an option which can be selected on bootup. But some departments such as Digital systems sometimes use custom devices which would take a lot of work to port to Linux properly.
Furthermore, Linux isn't as maintainable long term. Uni's are running apps coded 10 years ago without even a recompile.
We need to stop blaming Microsoft, and start working out WHY Linux isn't being accepted (despite being free). Trust me, it has NOTHING to do with Microsoft campaigning.
|
|
|
@Auzy How is linux not maintainable long term? You can image a linux distro across as many computers as you want and have them installed for as long as you want. How can an operating systems (such as windows or mac) which is no longer supported (by updates or otherwise) after reaching it's end of life more maintainable over long term than an os (such as linux or bsd) where you have access to the CODE and can pay someone to fix or update what-ever you need?
The only thing that makes linux a *little* harder to maintain is that the libraries actually get updated (imagine that) and that a few application developers don't stay up to date with those updates (python3 is causing a few issues there). But 1) Do you really want to buy an application that the maintainer isn't going to maintain in the first place? and 2) Since you get the code with most linux software, you can just hire an in-house guy to update it for you or do simple work-arounds (replacing python with python2 for instance).
BTW: @alix123 (spammer) you seriously spammed the SAME message to a single idea 3 times? That's just lazy, even by spammer standards...
|
|
Auzy
wrote on the 29 Dec 10 at 08:58
|
|
|
1) Even between kernel releases drivers need to be recompiled for starters. It's easy for you to say "oh, just recompile/update the drivers". Or "it's open source, just do it, or pay someone to do so". What if the original developer of a device is no longer available though? It is a lot of work..
2)A good example of the problem is hardware abstraction. Surely we can agree that's important. It all started with libhal only 5 years ago. Since then, DeviceKit was released to obsolete HAL, and now, udev is apparently replacing devicekit. So, I agree that I can't complain that Libraries aren't upgraded. HOWEVER, LIBRARIES ARE NOT MAINTAINED!!! Instead, they are replaced constantly by others. This is a MAJOR problem for developers. It's easy for you to say "oh, that's fine, just redevelop it", but it's actually a serious problem, especially when 13 year old libraries in Windows still operate..
3) There are still obvious forward compatibility shortcomings. Perfect example is shell scripts REQUIRE a hardcoded address for their #! declaration.
4) Upstart is another example. Whilst Fedora are working on a similar system backwards compatible with older scripts, distro's such as Ubuntu decided to just break everything. That doesn't excite developers. There must ALWAYS be a transition period. And it must be a proper one.
5) Finally, the biggest problem is the packaging system. There are too many. And it sys-admins see it as a sign of fragmentation.
3)
3)
|
|
|
If you want a system that doesn't need to be updated, then don't update it. If you want software that is maintained, chose software backed by organizations instead of individuals (firefox has mozilla, gimp has it's own, etc).
What would you recommend to allow linux to stay current but at the same time stay compatible with older technology. I have never seen a library in linux get replaced that was not outdated and in need of replacement. Sure, they could have patched hal, but that would have been ridiculously time consuming and ultimately pointless.
You mentioned 13 year old windows libraries still working. Have you *worked* with those libraries? They are unquestionably the most archaic api's I have ever had the displeasure of working with. It's not because Microsoft wrote them (some older linux api's are the same way) but because the software that relied on them 13 years ago have not had their back-ends worked on and thus Microsoft has been forced to include every version of every windows api in every windows release since about Win95.
Thins change, that is the nature of computers, trying to keep things "constant" is not only difficult but ultimately results in unmanageable code-bases and an inability to inovate. Just look at windows, they have been desperately trying to add much needed functionality to it since vista (proper compositing, effects, media sharing, etc) and they are STILL unable to truly revolutionize their desktop anywhere's near as quickly as MacOS (after their switch to the BSD modified kernel) or linux. Most of that failure is due to their constant need to provide backwards compatibility for application vendors who are too lazy to update PAID software that hasn't seen more than superficial changes in 10 years.
|
|
Auzy
wrote on the 29 Dec 10 at 11:48
|
|
|
Umm.. I don't think you are getting it.. AT ALL!
1) No system needs updating.. There are often productivity improvements though.
2) You totally missed my point about libraries. There is a LOT of stuff which needs updating (true). The problem, is that a lot of it, needs updating because the maintainers disappear.. Furthermore, rather than consolidating and trying to merge services, companies like conanical are creating new standards, without any transition periods. It's fine to create new standards.. It's unacceptable to do so without any consideration for backwards standards.
In regards to maintaining the api's, that is true. However, on the other hand, I own a copy of X-plane which doesn't install easily on Ubuntu (due to mounting issues) which is 1 year old. I own many games from 10 years ago, which ALL work on Windows still. So, sounds like Linux is great, if you want to maintain software. But it still makes Windows a cheaper target to develop for than Linux.
Also, in regards to paid software, lets be honest.. If Adobe ported Adobe Creative Suite to Linux, it would be unmaintainable. Just the most basic example is that anything that touched hardware (Premiere) would be a nightmare to maintain (especially thanks to messups like HAL).
And sadly, it's easy for you to call app developers lazy, however, some companies have limited resources. When developing complex applications with limited budgets, time can't be wasted rewriting everything every year. In regards to Linux innovation, Compiz was it. And I hate to tell you, but a lot of OSX innovation was stolen from elsewhere too (Time Machine for instance was a copy of Shadow Copy).
I think you are making the assumption that unlimited programmer resources exists, as long as it's open source. That's untrue. In fact, one project I saw gave up with Linux mostly because they found it was a huge burden on their project...
|
|
|
|
@PaddyLandau: I just read your comment again and I think I understand what you mean. I wasn't suggesting that it is the universities offering the discount. Yes, it is Microsoft, but as I said before as well, the ones that receive money from the government (therefore tax money) are spending some of that on software licenses (albeit discounted).
|
|
notgary
wrote on the 4 Jan 11 at 17:55
|
|
|
I'm not sure how it works in other countries, but in the UK, there are four way fr students to get their Microsoft software.
1) Buy it at full price from Amazon and be done with it.
2) Buy it from Microsoft at the student discount, with student status being verified using their education email address.
3) Avail themselves of the MSDN Academic Alliance, which at my University, constitutes a free copy of Windows 7 Professional, no payment necessary.
4) Use Microsoft Dreamspark (http://www.dreamspark.com/default.aspx), which is a service where students can get access to Microsoft's complete range of development tool for free. This includes
- Visual Studio 2010 Professional
- Expression Studio 4
- XNA Game Studio 4
- Robotics Developer Studio 2008 R3
- Windows Phone SDK Plugin for Visual C++
- and many more
The problem, at least in the UK, isn't one of cost, it's about mindshare. People know about Microsoft and they're likely to just go with them simply because of that.
I think solution #2 is going to be a localised one, with different consideration in different countries. Generally, I think publicity is more important to getting students (and anyone for that matter) to adopt FOSS products.
Solution #3 sounds great, but it sounds like the sort of thing that Canonical should not try to implement themselves, and rather some non-profit educational body should do so. I say this because having a single group, such as Canonical, focus on two many different fields will likely lead to sub-par results in all of them, where as if they just focused on the development of Ubuntu and it's variants, and allowed someone else to do the test/exam distribution (because that's going to require some serious infrastructure, including servers, developers and test writers/markers), then you get both groups focusing on a single field in which they can excel rather than spreading themselves too thin.
|
|
Auzy
wrote on the 5 Jan 11 at 08:16
|
|
|
You guys I think are TOTALLY missing the point still..
1) Departments such as Computer Science already often use Linux.. Mathematics MIGHT be able to use Linux (since apps like Maple are ported). Then again since the apps available for home users are way behind OSX/Windows, they wont use it at home.
2) There is no decent Photoshop/illustrator replacement.. Art departments cant touch linux (Gimp is years behind). Even if Linux did, there is no Aperture/lightroom alternative.
3) Any department that does video editing, has no final cut pro, or compositing software (shake/After effects).
4) Digital systems can.. And most already do..
5) English/language departments often need GOOD referencing tools such as endnote
Lets be honest though, at the end of the day, we need PROPER apps. We DONT need forks of gnome or faster bootup.
And I've spoken to many of my customers who agree. Canonical need to start focusing on stuff that's missing, rather than wasting resources on technologies that are good enough.
Trust me, most uni students would still require windows to operate efficiently.
|
|
|
Hi Auzy,
You have a point in that some areas, like multimedia creation, some free software alternatives are a little behind, although I think software like Cinelerra and Cinepaint (which has been used on big films in Hollywood) prove the opposite.
I think, however, that in the overall picture, if Unis were to switch to Open document formats to allow submission of assignments, etc., then everyone can use Linux.
Considering that Unis have historically had a huge hand in the development of software (consider BSD's contribution to networking) in all spheres, I think it is time they return to their roots and be the innovators again, rather than the victims of commercial software enterprises.
|
|
|
Auzy, by way of your examples:
I don't think any art department is obliged to teach in Illustrator Photoshop, just like no computer science department is required to teach programming with Visual Studio. If they want to teach the principles of vector illustration, they can use Inkscape.
In fact, no University is obliged to teach students to use commercial products just because that is what may be used in the workplace.
That is why I would actually expand on #1 and #2 and say that one should market the idea to universities that they can still teach the principles of just about anything using free software.
Not only that, but I just had another idea... I think I will add it as a solution here.
|
|
|
Sorry, but I just can't seem to leave this alone.
Auzy, let us suppose the Mathematics department of a University decided to use one or more of the several modeling and analysis tools available in the Ubuntu repositories (or elsewhere) for use in their curriculum. They would no doubt be able to use it. But, if they find any shortcoming, they or their students would be able to enhance the software, because they have access to the source code. In this scenario, everyone benefits, as opposed to the vendor of the closed source product.
Let's take it a step further: Imagine the Uni hosts their own mirror of the Ubuntu repository. Now any student can bring their Ubuntu laptop to campus and install any of the software, or connect to the Uni's network and do the same. Wonderful!
|
|
Auzy
wrote on the 9 Jan 11 at 01:52
|
|
|
Egh.. Ok.. Let me put it this way:
1)How often are Cinelerra and Cinepaint used? From what I can see, Cinepaint quite regularly. But I'm having trouble finding any real references for Cinelerra.
2) The last release of Cinepaint was 2 years ago. In contrast, the last version of Adobe Creative suite was released 1 month ago. That's probably why I'm having SERIOUS trouble finding any recent references to it's use.
3) We don't actually know much much they were used in any movie.. It's possible only 1 or 2 people on the team used them seriously.
4) Uni's might not have an obligation to teach what is used in the marketplace, but if you have ever tried to learn commercial tools such as maya (and have only tried more basic ones in the past), nobody is going to hire you. Because tools like Maya take WEEKS to learn, and even longer to become efficient with. Would you rather hire someone who was well trained in all the tools you are using, than someone who isn't trained in any?
Whilst I'd respect such a university, I wouldn't go! In fact, that was one problem with my university, all the business students were learning MFC, .Net and VB, whilst we were learning SML, C/C++ and PHP. Guess which students got employed the fastest?
5) Universities also have an obligation to use operating systems which just work. Linux doesn't. You only need to look as far as the new Sandy bridge processors, which require pages of procedures to get working currently in Linux distributions, whilst in Windows, it's easy. Everyone has hardware which doesn't work properly, and new hardware rarely works. So you want to force people to use outdated hardware in uni to ensure compatibility, and ensure they don't want to use plenty of commonly available hardware?
6) Teaching their students to get Linux and such installed at home will be a nightmare for their support department.
7) Ubuntu has at time obviously rushed technologies such as Pulseaudio in recent times.
8) It's difficult to recommend linux for mainstream use when the most popular distro's refuse to converge
|
|
|
|
I think part of the problem is a lack of textbooks. You need to have teachers have some sort of curriculum. This is not the whole answer, but something that needs to be addressed.
|
|
Auzy
wrote on the 9 Jan 11 at 02:36
|
|
|
But it's also application quality.. As mentioned, high-quality Linux software is VERY limited. And it's mainly because instead of improving applications, companies such as Canonical aren't allocating resources appropriately.
As mentioned though, this idea is implemented in departments already where appropriate (which at this time is mainly programming/science departments). However, a lot more work is necessary because it can even be considered for arts departments and such.
|
|
|
Hi Auzy, I think you need to qualify that second-to-last statement of yours: COMMERCIAL applications for Linux are in limited supply. Free alternatives, though they may not be as FEATURE-RICH as some commercial applications, are still high quality. It really depends on your definition of high-quality. If performance of data processing is the criterion, then Linux apps are often much higher quality.
Perhaps one option for Canonical to improve the applications in question is to generate revenue by selling support for some of these apps (the model many open source vendors use) while still making the software free, then using the revenue to improve the usability.
|
|
Auzy
wrote on the 10 Jan 11 at 21:55
|
|
|
Define high quality? Doesn't crash?
Lets be honest, if we were to compare maya, final cut pro, Adobe creative suite, iPhoto/picasa, and even Microsoft Office, the choices on Linux are less productive (Openoffice might be good enough for some people, but it is still missing features which are necessary in some environments).
When I talk about quality, I talk about productivity, and at this point of time, Adobe Creative Suite (which is used in almost every business) realistically has NO competitors.
|
|
|
Auzy, all you keep saying is that it is pointless trying to punt open source to universities because the free alternatives don't compare with highly expensive commercial products, instead of trying to think with us how we can go about meeting the challenge posed to us.
Please don't take this the wrong way, but your comments are adding absolutely no value to the discussion, other than trying to detract those with high hopes for Linux from the goal of making it the OS of choice for everyone.
|
|
|
|
Sorry, I realize that my remark in the second-to-last comment was perhaps a bit harsh. No offense was meant.
|
|
|
my 2 cents is a long term goal to target primary schools first. im putting it here and not as a solution coz it needs peer reviewed polishing.
primary schools are always fund raising for new computers when there old computers are adequate for the basic stuff they use them for. they also have students and (sometimes) staff that are not familiar with any os that would bias there attitudes.
start at the lowest grade and implement linux on computers at the next highest grade year by year so that by the time these kids get to uni microsoft will be forced to abandon its os projects and instead be writting unix softwware, thereby also solving bug #1. for some reason i feel european countries would be more likely to give a trial a go. perhaps it is possible some postgrad would be interested in doing research into how free software affects the quality and cost of education (proof of all those free software claims).
ps. im sure this was Shuttleworth's original goal anyway
|
Post your comment
|
