Slashdot Mirror
Survey On the Future of Open Source, and Lessons From the Past
An anonymous reader writes "Andy Oram reports on the quality, security, and community driving open source adoption. 'All too often, the main force uniting competitors is the fear of another vendor and the realization that they can never beat a dominant vendor on its own turf. Open source becomes a way of changing the rules out from under the dominant player. OpenStack, for instance, took on VMware in the virtualization space and Amazon.com in the IaaS space. Android attracted phone manufacturers and telephone companies as a reaction to the iPhone.'"
It's great to see Open Source used as a tool to help foster healthy competition where it otherwise may not happen. But it's also potentially bad if the Open Source path leads to worse results for end users.
Take for example the iPhone/Android comparison made. The iPhone took control away from the mobile phone carriers in regards to the device, allowing all iPhone users to see updates all at the same time. It also put a dent in the phone crapware problem. Android has done nether, suffering problems because devices can't be all easily updated. Google today announced that they will be updating APIs through Google Play. All because their attempts to update those APIs at the OS level failed due to carrier and device manufacturers holding up, or never providing OS updates. Google is only regaining control and providing better user experience on Android by becoming more closed, at least when it comes to how they deal with carriers and device manufacturers.
The facts don't support your claim.
Traditionally, it's been common to view price as a motivating factor, since open source software is often free. Last year, freedom from vendor lock-in was cited as the the primary goal.
This year, however, freedom from lock-in dropped to No. 2, while quality, which was in third place last year, was named the most important factor behind open source adoption. The availability of vendor support, meanwhile, is now a point of much less concern than it used to be.
"I've got more toys than Teruhisa Kitahara."
Quite the reverse. The whole point of a competitive market is that the best products get to the customers. In the absence of price competition, all FOSS projects can compete on is quality.
Projects which don't meet customer standards don't progress, though any of their innovation and effort which is of value is still shared. What the troll is trying to spin as a negative quite simply isn't.
Well, other than suggesting that Libre Office and the GIMP are like half-finished projects.... but yeah. His first sentence I agree with completely.
File under 'M' for 'Manic ranting'
So fork GIMP and call it something else. If the name is the only thing stopping corporates from putting money into it so they can break Adobe's monopoly, that would seem to be easily solved.
The larger argument about "consistency", as if corporate software is consistent in quality, is just too ridiculous to even argue.
People willing to spend $2000 or $600 on a bit of software are very resistant to change. It doesn't matter what license the alternative uses.
The problem isn't the "quality" of Free Software alternatives but the fact that NO alternative of any sort will be considered acceptable because software consumers tend to have a mentality fixated on single brands even when the data formats involved don't have any inherent lock-in.
Shills with no money but lots of free time to post on web forums help contribute to the sense of "single brand ineveitablity".
A Pirate and a Puritan look the same on a balance sheet.
All that proves is that Free Software is more transparent.
A Pirate and a Puritan look the same on a balance sheet.
I'll disagree, there is no equivalent or better alternative in the open source world to the proprietary products I mentioned, the job simply cannot be done in the open source world. But for other softwares, say an operating system or a browser or general purpose scripting language, the open source is superior.
"For every OpenStack there are craploads of half-finished projects that are basically in a perpetual beta stage. Documentation is spotty, features are spotty"
What makes you think that OpenStack is -as of now at least, any different?
All that proves is that Free Software is more transparent.
Fucking this. Right here.
I'd argue there are orders of magnitude more half-finished projects than fully developed ones in closed source environments, it's just no one sees them. At least I know I have about a dozen to one ratio of crap projects versus stable, releasable projects in my own folders at work. And I'm not a full-time programmer. I certainly know that the releasable, stable projects started out as half-finished crap.
The open source community might do well to have a major hub adopt an easy to use TRL standard for devs to mark their projects with. Perhaps reviewers could even agree/disagree with the developer's marked rating. Perhaps certain trustworthy users could even function as a third score (Dev TRL, Registered Users TRL, Critics TRL) or something.
Well, there is a lot to respond to here, but I'll just address the question of whether the trend will continue. The article itself seems to refute your statement. The survey indicates that the open source trend is increasing, so I think the more reasonable conclusion to draw is that it is likely to continue for some time.
yea, but there's just as much crap closed source products as well. The only difference is you can see the garbage in an open source product. My employer has gone from being very annoyed with having to deal with open source licenses and trying to get the whole idea of "it's free, we can't negotiate the license, there's no support contract" though their approvals process... to now just having a check box for which GPL version it is and an automatic approval process. It's great now. Granted we're limited in the scope of what we build with a GPL product. If we're building something that hundreds of people will eventually depend on and we have no way to back out... then that project is going to get a lot of scrutiny. The one good thing about closed source vendors is you can sue them if shit hits the fan.
You entirely missed the crux of the summary. Each company knew they could not, individually, supersede the established competitive advantage of the most successful player in their industry. However, bandied together in cooperation, they COULD forge a competitive advantage and undermine the player's supremacy. Better to collaborate on an alternative than concede and pay out millions to the player.
It's like Zulu uniting tribes against The British Empire. Or Attila the Hun uniting tribes against Rome. Or Genghis Khan against China.
If your business model needs you to the be the sole owner of a competitive advantage, and you are never able to achieve that advantage alone, then you have no business model.
"Love heals scars love left." -- Henry Rollins
China has really only gone from poverty when they abandoned communism. Look at China during the pro-communism era (Cultural Revolution) where historical sites were desecrated, many were killed and starved, and political and religious freedoms were suppressed.
Compare that to Hong Kong which has been (mostly) capitalist under British rule and it was much more prosperous (and still is!) than the rest of China.
Today, we can't even accurately tell the growth of China due to manipulated statistics, but China is undoubtedly in a bubble with manipulated statistics and fake construction ( http://www.news.com.au/business/china-building-mega-cities-but-they-remain-empty-sparking-fears-of-housing-bubble-burst/story-e6frfm1i-1226611169281 ) producing decent numbers but no wealth.
Taxation is legalized theft, no more, no less.
You bring up two points that are worth addressing. There's some truth to both, and there's good news on both - they are solved by using open source in a way that makes sense, not thinking it's exactly the same as proprietary software, except you get the source code. It's kind of like saying that dogs are better than cats because your cat won't play fetch. True, cat's don't play fetch, instead they play with laser pointers.
You're right, you can find lots of beta level OSS. Both free and proprietary software have betas. With proprietary software, you're not allowed to see the betas (unless it's Microsoft, in which case your new computer comes preinstalled with Windows 8 alpha.) With free software, you can choose the beta version of a mature project (Fedora), the stable version of a mature project (Red Hat), or the beta version of a new project (FuSe). They are all available. That means you'll want to look at the status of that version before making a major commitment to it. Don't install FuSe if you want a mature system, install Red Hat. It's actually cool that you CAN choose FuSe, or a development kernel, if you want some new feature that's in development and not yet rock solid stable. You do want to check though, and that's why Sourceforge shows you right up front how much activity the project has, the version numbers, user ratings, etc., so you can choose maturity vs. bleeding edge, etc.
You also mentioned documentation, which is sometimes important, and is actually entirely separate from the quality of the software. True, the programmers of OSS have less incentive to author well organized, newbie friendly documentation in the style you're accustomed to, unless you use a certain trick. There's actually MORE in-depth documentation for OSS. Every change to the software and the design decisions are normally documented three times: on the -dev mailing list, in git/svn, and in bugzilla or similar. If you have a question, you can email the list and the authors of the software will answer you, assuming you ask a Good Question (see ESR). So if you want to really understand how something ticks, you can find lot more information about how Apache works than how IIS WORKS, for example. That's not too newbie friendly, though. For comprehensive, newbie friendly guides, you need one of two magic words.
HOWTO is the first magic word. Google _____ ______ HOWTO for any OSS topic and you'll probably find the documentation you're looking for. If not, the second magic word is "book". I work on a OSS project you've probably never heard of, Moodle. Moodle isn't a high profile project, yet Barnes & Noble has EIGHTY listings of Moodle books. That's EIGHTY versions of the comprehensive documentation you're looking for. (Could be 40 different books, B&N may have duplicates listed.) I know, you're shocked. I just suggested BUYING something related to open source software. I know it may seem strange, but compare $500 for a Microsoft solution versus $22 for the book to go with the free software option. I'll take $22 over $500 all day long if I really need 150 pages of illustrated documentation.
So you're right, OSS projects don't prevent you from downloading beta quality code. And dogs don't catch mice. Consider this post as "Intro to Cats, a Guide for Dog Lovers".
The problem with GIMP is that it has a horrendous usability problem that seams to increase per release.
Yes, GIMP is a stupid name (even as an acronym) ... but a name alone doesn't make or break a product. USABILITY is the #1 factor in making a software product successful.
The survey, as most open source articles, studies, etc. ignores the elephant in the room: open source leads to loss of competitive advantage for companies. I know I likely won't ever use open source software to run any critical parts of my business, because part of my business model is having a competitive advantage through better software than my competitors.
OK, so then I'll just start at the closest open source thing to yours, add less work to get even more of a competitive advantage than you have. Me and a few other competitors will even split the bill, even on maintenance. Leave you in the dust. Good luck with that.
Protip: Your ability to configure the bits is what's valuable. Free and Open Source software means more folks configuring bits in one place. Also note: Even with GPL'd software you're free to use the software and modify it internally so long as you don't distribute it.
OK, so then I'll just start at the closest open source thing to yours, add less work to get even more of a competitive advantage than you have. Me and a few other competitors will even split the bill, even on maintenance. Leave you in the dust. Good luck with that.
Hasn't happened in my industry (retail), and it's not likely to. Heck, there's not even a single open source package that is at all competitive in the small to mid size point of sale market. Correct me if I'm wrong, but that's a pretty big market that somebody would have taken advantage of, if your theory held any water.
I don't respond to AC's.
What you're want is not the goal of a license -- what you're wanting is the goal of a written (or in some states/countries, verbal) contract/negotiation. Understand the difference between what a usage license is for and what a financial contract/negotiation is for. Understand they serve separate purposes.
My advice to you would be for the "someone" to use the 2-clause BSD license for the licensing, and a written/signed agreement between you and said "someone" that discusses financial reimbursement (percentages, rates, baseline payments or maximum payouts, etc.). The latter doesn't have to be complex, it can be as simple as you want it to be (literally hand-written on a piece of paper if you want), though I would strongly suggest having a lawyer review it for any "catches" that could screw you financially in the end (e.g. someone 5 years later smiles and says "you said 15%, but didn't say of what, so I want 15% of all your company's assets"). Otherwise if you don't like the 2-clause BSD license, consider the WTFPL (yes I'm serious).
Both the 2-clause BSD license and WTFPL are easy to understand, and provide actual freedom of choice/action; while Richard Stallman's desires might be founded on good intentions, it may or may not apply to every situation. The GPL is a very "idealism-focused" license (in the sense that it's hoping to change the world through a borderline viral belief system), while things like the 2-clause BSD license and WTFPL are more pragmatic.
You would be surprised how many software authors choose the GPL simply because "they heard of it somewhere", and likewise how many software authors do not understand in full what the GPL says or mandates/requires. It's quite shocking, honestly. I would say "Don't people read the license they choose?!" except the licenses are written by pretentious pricks and are very hard to understand, especially for non-English speakers (I had to deal with this last month, where a Brazillian colleague was considering the GPL but didn't understand it given how its worded). The 2-clause BSD license and WTFPL are pretty black-and-white -- once again, KISS principle wins.
Sure, no one in the whole world can get the job done in GIMP, it's like Photoshop has some magick properties that makes it the only software that can alter some imagebits... I'm not saying that The GIMP is better than Photoshop and I do recognise that Photoshop is better, it's just that it's not that much better and OP is right, the major problems most people have with GIMP is that it's GUI isn't a 100% clone of Photoshop.
A generic license like that is basically pointless, because it has to define what "revenue" actually is. I bet that Google could prove that they make no profit if they wanted to. In fact, they do prove that they make no profit, which is why the poor souls don't have to pay tax.
If you want this form of license on software that is GPL, then write to the authors and ask them. This will involve a period of contract negotiation, and they will probably only be bothered if you can convince them that you are likely to turn a profit. And away you go.
The bottom line problem is not that GPL is bad. It's that you don't have any money. Over the last twenty to thirty 30 years, we have transformed our societies to one where the market rules us; you don't have any money, well, tough.
The GPL is hard to understand, because it is quite long. This is because it is written to have legal meaning. It's the legal system(s) that is at fault here, I think.
I'll go one step further, from my observations of the typical arguments seen re: GIMP vs. PS:
1. The biggest problem seems to be the name. Silly, but what can I say? I do agree that the name sucks, but oh well.
2. Next would be the lack of a (default) MDI-type interface, as horrible as that style is. Yes I know GIMP has that option now. Most people don't know this.
3. It lacks obscure feature X or Y which is useful to a certain percentage of users, like the Libre Office vs Word arguments
Personally, I have no problem using GIMP on Linux or OS X (I don't use Windows except at work to launch PuTTy), but then again, any image adjustments I do are simple resizing, rotations, color corrects, sharpening, format conversions and other simple tasks. Sure the name is stupid, but that doesn't bother me.
What mystifies me is *why* there is no OSS PS clone; it's not like this is rocket science. Most image processing algorithms are well known; a simple web search will find them, as most come from papers submitted to SIGGRAPH (sp?) over the years. In addition, it seems like it would be a fun and interesting project, unlike say accounting software, so that kills the 'boring' argument. And last, this would probably be *the* most popular piece of free software, end of story. Whoever wrote it would be a frickin' rockstar.
Going back to my original points, the biggest issue w/GIMP is probably the fact that PS is one of the most widely illegitimately distributed pieces of software. If it wasn't so easy to get a copy off the 'net, it's usage would be a small fraction of what it is today.
> know I likely won't ever use open source software to run any critical parts of my business, because part of my business model
> is having a competitive advantage through better software than my competitors.
Then you're doing it wrong. Specifically, you've fallen into a black-or-white view and forgot that 90% of cases are gray.
Surely your company didn't write it's own mail client, you use something like Outlook, Evolution or Claws because there's not nearly enough competitive advantage in having your own email client for the ROI to make sense. You don't write your own word processing program for writing memos and letters. That's one extreme.
At the other extreme, you may have ONE piece of software which does that thing that sets your company apart. You might open source or outsource your payroll, but Google would never open source their search algorithm. Ebay may open source or outsource their forums, but the core of their auction system is the core of their business, so it's 100% proprietary. That's the other extreme. So the two extremes are a) commodity software where you should use something off the shelf and b) your core competency, your true competitive advantage, key business secrets, which should be well closed.
90% of what you use doesn't properly fit either extreme. That 90% in the middle is where the ROI is best by CUSTOMIZING existing software. Ebay doesn't write it's own web server, they customize Apache. To handle petabytes of data, Google customizes open source storage stacks. For MOST things, being completely dependent on a vendor for updates, support, etc. is at least a risk, so committing to of-the-shelf proprietary software instead of an open system you'll be able to customize if needed is a mistake. At the same time, building from scratch is a huge waste of money for most of the software you use. Most cases fall in the middle - use what's already available rather than writing your own, but not by becoming dependent on a third party vendor who may pull an Adobe and decide to stop selling their desktop software, instead offering it only as a cloud based monthly service (or who may go out of business entirely). Open source fills this huge middle perfectly.
Think I'm wrong? That means you think Ebay and Google are wrong. Over half of the world's largest companies, the Fortune 500, are known use open source software. Are you REALLY smarter than Ebay, Google, and all the other multi-billion dollar companies? Is that proven by the fact that you're more successful than they are?