... They probably did so because it would shift the focus away from the KHTML they know and love and more towards the more realistic (but messier) WebCore, which they don't seem to want to do. The KHTML team doesn't even seem to want many of the changes. Apple makes a product, and they don't care if they break small things to make deadlines.
Can anyone provide evidence of Apple's work on KHTML being messy / hacky / etc? So far, all we've heard from the KDE folks is that they didn't like the fact that Apple's patches were monolithic and thus very hard to integrate. What is the actual quality of Apple's WebCore code? I started posting earlier but then realized this whole "WebCore is messy" seems to be unsubstantiated. Somebody find some proof!
I think one of the strengths of Open Source is that developers are not under economic presure to deliver it yesterday. They have usually taken the approach of getting it right. I think this means products are sometimes longer to market, but its a trade off.
I'd like to point out here that the only reason Open Source development usually takes longer is that, as with KDE, it's largely a volunteer effort. Pay those same high-quality, principles-first developers a full time salary, and I'll bet development is just as fast as any corner-cutting proprietary shop. Open Source needs commercialization but it needs to be done properly.
What Apple did was simply fork KHTML because they insisted upon absolute control. If they had instead hired / contracted with the original KHTML developers, none of this mess would have happened and everyone would have been better off. To blame the volunteer KHTML developers for not accepting low quality patches to their hard work is asinine. The KHTML developers had put an enormous amount of hard work into making their codebase clean. I know from my own experience working with / managing Open Source projects that low-quality patches from outsiders almost always come back to bite you in the future. I blame Apple for valuing control over quality. Timeframe was not the issue.
KDE has chosen a better infrastructure than Gnome... But Gnome is better in terms of the software they developed.
KDE's infrastructure allows for a modern, well-integrated desktop. It is an example of object oriented principles being properly applied in real world development. Everything is richly context sensitive and there is a high degree of code modularity and re-use. To give a specific example, application file dialog widgets in KDE use the same objects as the Konqueror file manager component. Unlike in GNOME, I can manage files, open a preview pane, access network resources, etc. within the file dialog. KDE has many other similar qualities that GNOME lacks. From a real-world usage perspective, KDE is highly superior to GNOME and it's no surprise why most people running Linux today use it. I would have to say the same for KDE applications -- they're generally much richer because they re-use the superior core KDE components. Examples: I know of no equal equivalents to K3B or Amarok as GNOME/Gtk apps. I also find KMail far superior to Evolution and Kopete superior to Gaim.
KDE tries to look like Windows too much for my taste. Gnome is inspired on the same ideas of Mac OS, something more elegant.
KDE doesn't really try to look like anything - it's a product of what seems to work and what users are asking for. I certainly don't see much similarity to the look/feel of Windows. (Yes, I still run into Windows a bit on the job so I'm still quite familiar..) GNOME is (now) designed upon a set of HCI guidelines developed by a handful of folks who sat illiterate users down at computers, asked them to perform simple tasks, and then decided that this must be the way interfaces should be designed. Obviously everyone else (aka. real users) don't know what they're talking about. The result is agonizingly simple, inflexible, and feature-bare software. That's not to say there isn't some great GNOME/Gtk software but the desktop environment itself is the pits. It's not MacOS X either.. it's more like Mac OS 7 or 8 in feature set / richness. Hardly "elegant" by any standards.
Every single business would go for the $80,000 propriety package. Firstly, businesses buy from other businesses, not from college students programming in their spare time, none of who want to have responsibility and their ass on the line.
Obviously you didn't even read my post because I said professional, commercial development not college students. But yeah, I agree: college students and hobby hackers are usually useless to produce quality code.
Secondly, no business cares that 12 people 'could' pool their resources together a year from now. They care about what WILL happen, and put that into a license and buy from a company whose whole job it is to support the product.
Wrong. Around 70% of all software is still developed in house. Collaboration between companies with similar needs makes perfect sense. And again, you are making silly assumptions based on taking my post at face value without actually reading. I never said anything about "x number of people" pooling resources. Get the stupid hobby hacker Open Source myth out of your head.
Thirdly, businesses expect tech support and future existence.
Which collaborative open source projects guarantee whereas proprietary software does not.. (And yes, as an IT admin, I've typically gotten much better support from professional OSS projects than from most proprietary software companies.)
Fourth, a company with employees sure shows more dedication to the product than a sourceforge archive.
Back to the stupid hobby hacker Open Source myth..
Finally, no business does things for the lower cost, they do things for the best impact for the business at that time.
Good businesses think long term as well. In the long term, a transition to Open Source solutions makes sense more sense. If a solution does not currently exist, there is often a good business case for investing in its production and deployment over a given time period.
I think you're confusing capitalism, with the the idea that govenment agencies are required to take the contractor with the lowest bid. Businesses do not operate on such a stupid premise.
Cost savings is not the sole reason many businesses are exploring / pursuing OSS. It's just an added benefit that comes with the territory. The largest benefits of OSS are control and flexibility. Of course, when you're talking $40k vs $80k, you're talking about a small company - so cost itself may be a much larger factor.
At the end of the day, people don't pay for what they value. They pay for what they have to pay for.
Exactly.. and with OSS, what people sometimes have to pay for is the labor to initially create it. That may come indirectly through support contracts or directly through development contracts. At the end of the day, you go with what does the job for the lowest cost. If a proprietary package will cost your business $80,000 to license but it will only cost $40,000 to contract required development of a mature OSS project, which is a better deal? And what happens when dozens of similar businesses are able to pool resources to make features happen? (This is why I have always advocated that most OSS should be professionally and commercially backed.. it allows for ad-hoc business alliances and incredible economies of scale)
I'd rather see this fixed in 1.0.4 or ASAP. It can't be that difficult to implement.. it's just a simple update manager and it only needs to work in Windows anyhow. Linux distros have their own update mechanisms. (although a popup reminder might be good.. "an update is avail.. [howto message customized by the distro here]")
While "many eyes" will eventually help Firefox to become almost perfectly secure, it is still a very young piece of software. The Moz. team really should have anticipated future problems and built a proper update manager *before* releasing it unto the general public. Being less buggy than IE is half the marketing drive. This is not a reputation that can stand the risk of being tarnished.
So, if anyone on the Moz team is listening, here's my vote to fix the update manager immediately instead of waiting 6mo+ for version 1.1. I don't even use Windows anymore, but I want to see my Windows-using friends continue to use Firefox.
What they really need to do is stop trying to emulate Microsoft Office. You'll never make the MS Office killer by making MS Office.
That's correct. But what we need is not another new approach to standard word processing. We need to look past word processing altogether. When you really think about it, it's a totally outdated paradigm in the modern, web-connected world. Knowledge workers shouldn't be doing their own layout and formatting unless it's actually DTP. It's time for web-based document production. Forget WYSIWYG.. that's for the layout people. The vast majority of document production can be handled by lightweight formatting via HTMLAREA style web forms. All the final formatting can then be handled by a template engine. (such as to comply perfectly with company guidelines, etc.) Document workflow / approval is all handled in a uniform fashion through the same intranet interface. Imagine the potential for automation: you update the company newsletter, somebody else proofreads it, your boss ok's it, and then it automatically gets processed into emails, letter head mailings, and the public website. No swapping around documents via email or shared folders. No manual preparation of each output form. This is the Windows/Office killer, not OO.org. (Oh yeah.. and since it's web based, people can work from anywhere.. IT at its finest)
Here's my take on it: Microsoft Research was founded not as a way of performing valuable research for Microsoft, but as a way of preventing smart people from doing valuable research. By locking up the industry brainiacs, Microsoft can virtually guarantee that no one will come up with the technology to challenge them.
Or they're just astroturfing all possible future paths of innovation with patents. I think that's a more logical conclusion given the circumstances.
Well M$ did spend good money on research (billions of dollars) and lending technology so that they can get some of that back does not seem so bad.
But then you ask.. just WHY are they spending so much money on this "research" when it is plainly obvious from the history of the industry that most software related innovation happens automatically and incrementally. Let me give you a hint: MS is one of the few players that really truly supports software patents.. and it's because they're one of the only ones that benefits.
So no, MS should not be allowed to profit on the vast majority of the "research" coming out of their labs. It's not because good ideas aren't being generated. It's because those same good ideas would have come out naturally without one company spending billions to force them out before their time. Do you see what they're trying to do? In the face of competition from Open Source and (moreso) an industry quickly shifting to vendor-neutral standards, they're trying to monopolize innovation itself. Why do you think they suddenly set up all those cheap overseas research labs? I don't see any dramatic improvement in their products. No, they're basically astroturfing all possible future directions of innovation with patents in attempt to make it impossible to compete without them still getting a piece of the action. That's not capitalism. It's unethical business practice.
Software patents are wrong and this is a perfect example of why. Hopefully most of the rest of the world will maintain a better grasp on sanity.
However, it's pretty damned good for the times where you want a machine to parse your config files and present them in a nice GUI format.
I think there's an even better solution than either straight XML or custom config schemas: a compromise that uses the best of both worlds..
As per the Elektra (http://elektra.sf.net/ project concept, store all configuration values in plain text key-value fashion but use the filesystem as the database. (Think of the/proc and/sys virtual filesystems) However, in addition to this, store metadata about the config hierarchy in a standardized XML schema. This way, you can quickly and easily read/edit by hand, software doesn't have to spend time parsing XML on load, the security is more granular, and yet GUI tools can parse the config in a standard fashion, gaining all the benefits of XML driven configuration.
So I think what Apple did rules. They didn't ask requirements from anyone, they had a flash of inspiration, created something awesome that works good for them, and then said 'hey, if you want to use this, go ahead!' Apple never considered asking other Unices because they wanted something that worked, and also wanted to deliver too.
The problem is that most members of the "old school" Unix and even OSS community no longer have any flashes of inspiration left in them. They've been using the same crufty ideas for the last 20-30 years and as a result they can't think out of the "old school Unix" box any more. The fact that Linux and the BSD's haven't really evolved much in the last 15 years compared to OSX should be ample proof. I think it comes down to this: we need more fresh blood.. developers who are willing to look at the same old problems without a preconceived notion of what the solutions must look like. Old school folks who refuse to open their minds to radically new ideas need to just hang their towels and pass the torch. The XFree86 -> X.org transition has been a perfect example of this. The same needs to happen with all the fundamental Unix tools. Apple's ideas are interesting but they could be made even better by an OSS community actively open to change.
There's a peculiar way that we could possibly exploit the long tail phenomenon to severely cripple the "blockbuster" business model: flood the P2P networks with fake copies of all the most popular pop songs. I'm not talking about making mp3s of white noise or fart sounds. I'm talking about legally sharing indie music renamed to look like the titles of the pop crap. (preferably only quality and similar style indie stuff..) So when average users download the latest "britany spears" or "dmb" song, they instead get exposed to something else. If it's good enough, many people will actually stop and listen to it. Then, at the end, add a brief DJ-style plug of the band and song they just heard.
Possible benefits: - more exposure for independent artists - less exposure for pop music rubbish - anyone who gets sued based on the filenames can stand up to the RIAA in court knowing that there is no valid evidence against them - it would make P2P less useful for swapping RIAA content, likely reducing the number of lawsuits and simultaneously restoring P2P to what it was originally intended to be: a channel for distribution of freely licensed content.
Go to any trading floor or any advanced financial institution and they will rely heavily on Excel.
In this particular usage, Excel is basically a really fancy calculator and data analysis tool. However, I've seen absolutely disasterous uses of Excel as a replacement for a proper database application. I think that's what most commentors here are referring to. No spreadsheet is the right tool for the job in this case -- whether made by MS or not. Anyone who puts together a hack-job solution using Excel/Access or OpenOffice Calc/Base instead of buying or building a proper database solution is a poor IT manager at best. I therefore find the premise of the book meritless, though maybe it's a good "way out" for those who have already started down the wrong path.
Much of the "real world" has been stuck in a costly technology tar pit for the last 15 years or so. It's not like some sort of success story. Thanks to people like you, who obviously don't care about progress, the vicious cycle continues.
1) You have the following data sources: Oracle, Access, CSV Files, Microsoft SQL, and IBM DBase
OK.. stop right there. If those are the data sources in use for mission critical applications at some company, then the person making the core IT decisions has failed miserably. Now, I'm not saying that there's necessarily a better option than using Excel in this case, but the only reason Excel would be used is as a last resort. Starting from scratch, you would never ever use Excel to write applications or report generators. Ideally, the primary database software should be good enough that the use of spreadsheets can be banned altogether, with the possible exception of using them as a data "scratch pad."
I think this breaks down somewhat when you consider the importance of the developers. In this particular case, the purchasing company not only got the code, but the lead guy who created and/or managed the code.
Remember that the lead guy could always leave the company and maintain the fork if things went south. And if not, well, what's to say that he would have continued working on the old code (without pay) otherwise? What I'm saying is that it's a necessary risk to take. All major OSS projects need to have commercial backing so that the core developers can work full time on them. Ideally, the core developers run the show as well. Ideally, there are no proprietary "add-on" products necessary to sustain the free project. Ideally, all resources go towards developing and supporting 100% open code. So this particular example is not "ideal." However, as a user of JasperReports, I would much rather see this happen then for it to remain just a hobby project of one man. As long as JasperReports continues to be developed at least at the previous rate, this is a clear win for the Java OSS community.
That is a very good factoid that I had not noticed when I read TFA. It definitely raises the potential for the invention in terms of practicality.
At very least, it's probably worth the cost of improving waste water treatment plants. Energy and resources are going to be used regardless, so if we can get something extra in return, why not?
It has been over a decade since the last innovative new spreadsheet.. Time for something new.
Indeed.. time for something new. Spreadsheets, the poor man's database, are finally beginning to die out. Good riddance! Make way for RAD tools and web database applications! Any organization that stores important data in spreadsheets has something severely wrong with their IT department.
..What was the huge need for expensive MS Office again?
First: Will students really need to know OS solutions in the marketplace, or do we see the trend as continuing with windows/office in the workplace? This is a real argument. I personally feel that OS Solutions are going to be part of the marketplace in future years and that the time is now for Education to start to offer OS solution classes...not replace Enterprise solutions - but offer them side by side.
The inherent problem with any classes like these is that the technology changes so rapidly and the tools actually used in the real world are so diverse. Sure, everybody today needs to know how to use a basic word processor and spreadsheet. But beyond that, where do you even start? I see business students trying to learn Access and FrontPage. Will they ever use them? Hopefully not!! I think a more appropriate approach to technology education is: "will students be able to properly handle whatever the marketplace throws at them?"
The second question, and the most important, is rather or not we can find the teaching materials (text books, etc) for the classes and rather or not the classes can be accredited. All the OS solutions in the world mean nothing to education standards if we can't accredit them. That is what I am trying to do...find text books and all the infornmation needed to accredit courses that would teach OO and Linux (as well as other basic Open Source Solutions). So far, I have not had a lot of luck in that department.
Why not get together with some profs and start writing your own textbooks? No harm making a little extra cash while scratching your itch.:) You can be certain there's a market today. Probably an eighth of the textbooks I used in college were co-authored by local profs and some weren't even that widely published. As for "teaching Linux" you probably want to avoid getting bogged down with details, like anything sysadmin related, since these are so subject to change. But familiarization with popular OSS tools and desktops can't hurt. (KDE, OO.org, Gimp, etc.) And don't bother trying to teach business types any CLI stuff. They're usually not brainy enough to handle it and it just gives them a bad impression of Linux since they don't know any better. (ie. "Oh.. this is like that old DOS crap")
I'm sorry, but the only people who really NEED to know how to use a computer are those of us who keep them working.
You know the saying, "Give a man a fish and he'll eat for a day. Teach a man how to fish and he'll eat for a lifetime." Well, teaching people how to just barely get by with using Windows, Word, etc. is an example of giving them a fish. Sure, it's the most convenient option for you, but it doesn't really help them in the long run. Change the technology and you'll have to completely re-train them since they never really understood what they were doing to begin with. (And I doubt you are closed minded enough to think that technology never changes.)
And how do the said businesses know that OpenOffice will be around in 5-6 years? Will they be able to get support or training for it? Will there be up-to-date documentation? Maybe, maybe not.
There you go repeating the standard senseless FUD again. Yeah.. you know, because large, popular Open Source projects just randomly decide to disappear one day. And you know how awesome the free support and training is that you get from MS is when you license their products. These arguments are all bunk. All business decisions involve risk and there's very little risk in going with OpenOffice for those whom it meets the needs.
Will Microsoft be around in 5-6 years? A most definite yes.
Will MS Office still be popular in another 5-6 years? A most definite maybe not.
It would take Armageddon to do in a flagship MS product line. You must be having delusions of OpenOffice granduer to think that.
Dozens of "invincible" companies and products of the past have disappeared as markets have changed. Do I think OpenOffice is a long term solution? Nope. (but neither is MS Office) However, OO.org is a viable intermediary until next-gen web-based technologies are more readily available to replace it as well.
I'm sorry, but in general most open source projects that are supposed to be replacements for commercial software just don't cut the mustard.
Open Source is only now reaching the mainstream. It is quickly emerging from its non-commercial, hobby-hacker past. Don't expect the future to be anything close to what we've seen in the past.
They don't look as good, they don't feel as good, the support isn't there, there's no guarrantee of reliability, and just doesn't look good to a business and most individuals.
Look and feel are almost entirely relative. Anything unfamiliar doesn't "look and feel" as good as what you're familiar with. I could say the exact same thing about Win/Office. I hate the basic Windows GUI compared to the much richer KDE interface. "Support isn't there" is simply a lie. And there's no guarantee of reliability in most commercial software either. Try reading some EULA's..
the vast majority of Businesses use what everyone else is using, and that's Windows and Office. Scream, cry, kick all you want u can't fight reality. Kids should be learning what the business world is using, putting on your resume "Redhat and Open Office experience" doesn't look that great to the average business running Windoze & Office.
First off, this has absolutely nothing to do with Linux. Anyone who can use Windows can use KDE. I've seen dozens of examples of this -- especially in education. So the OS is irrelevant as it is an implementation detail. It's the apps that matter.
As for OpenOffice, anyone who has trained on it can honestly put both MS Office and OpenOffice on their resume if that's such a big deal right now to clueless employers. There's virtually no difference to the average user and nearly all business users are such. The exception is specially trained secretaries who have been sent through dozens of expensive seminars on fine details of various software. This is an extreme rarity.
And BTW, do you know why businesses will not be running Win/Office in the future? It's not because Linux/OO.org will become perfect clones. (although this is an intermediate step for some) It's because the word processing and personal desktop paradigms themselves have seen their day. Businesses that stick with the old paradigm technologies will be left in the dust by those who innovate in the area of workflow efficiency. We're on the verge of a massive upheaval of the status quo and the ushering in of a new generation of business computing technology. It's been 20 years since that last happened, thanks to the effect of proprietary monopolies on true innovation.
I've heard that idiotic argument used before. The fallacy is that kids should be taught how to use specific applications that they'll find "in the real world." Wrong! They should be taught about how to use and truly understand computer technology. By the time kids in high school now are graduating from college and getting their first real jobs, MS Office may well be a thing of the past. Honestly.. do you think in 5-8 years that OpenOffice.org will not be just a tad more attractive to businesses? (or some other project if not OO.org) Or how about modern web-based document management / production systems that eschew the silly, outdated "word processing" concept that keeps today's businesses tied to inefficient workflows and excessive paper waste.
And before you say, "Yeah, but what will they need to use in college?" consider what you used in college. Was there anything that OO.org in its current imperfect state could not handle perfectly well? Typing essays and reports? Including a simple table or chart of your chem lab results?
The problem with most schools is that they focus all their energy and resources in providing students with the "best" facilities, equipment, etc. and then miss the whole point of properly educating with an eye on the future.
Supporting all of them with all the various potential support problems with varying library support would be a nightmare, even with _only_ 4 different linux distributions.
In reality, most binary software runs without a hitch on most distros. The handful of support issues that do arise is no worse than the multitude of Windows-related support issues companies currently deal with. The whole situation is far far exaggerated.
Large businesses want to run Oracle for instance.
Businesses with requirements complex enough to actually need Oracle (ie. not just buy for the name) are increasingly rare. Any that do are going to run it on officially supported RedHat or SuSE as a standalone server. In other words, the choice of distro is entirely secondary to the choice to use Oracle. I'll leave aside my commentary on how high-end databases are usually a band-aid for poorly written applications / middleware...
Open source software is no more going to kill off closed source software than the internet is going to kill off printed media.
Don't bet on it. In another 10-15 years, the only "closed source" software will be application service providers -- and even these will be built using OSS. It makes too much sense and the momentum shift has already begun.
And yes, the internet will kill off printed media.. because it'll be downloaded to electronic paper instead.:)
Because not everyone wants to be an open source software company. If linux is ever going to be used by a business, a regular end user, etc it has to be able to support closed-source programs. That means binary compatibility so a software maker doesn't have to support 15 different compiles of the same piece of software for each distribution.
First, any business running Linux is reasonably going to be using only one of 3 distros: Fedora/RedHat, Debian, or SuSE. All of the rest are special purpose distros which can be safely ignored. If you're running a proprietary shop, that extra hour of compiling/packaging isn't going to make or break your operation. Second, if you really want cross-platform software, you should be using Java and/or developing web applications to begin with. Binary-compatibility-driven desktop platforms are on their way out. Businesses are increasingly looking for ways to disconnect their IT needs from vendor and technology lock-in. Third, the availability of closed-source software for Linux is not a requirement for its success in business or even for home end users. The gaps remaining (special purpose software) may just as likely be filled by Open Source software as proprietary. It all depends on how well Open Source communities leverage the business world and vice-versa. There are plenty of level-headed (non dot-com) business strategies that have yet to be explored by Open Source entrepreneurs.
Slashdot Mirror
... They probably did so because it would shift the focus away from the KHTML they know and love and more towards the more realistic (but messier) WebCore, which they don't seem to want to do. The KHTML team doesn't even seem to want many of the changes. Apple makes a product, and they don't care if they break small things to make deadlines.
Can anyone provide evidence of Apple's work on KHTML being messy / hacky / etc? So far, all we've heard from the KDE folks is that they didn't like the fact that Apple's patches were monolithic and thus very hard to integrate. What is the actual quality of Apple's WebCore code? I started posting earlier but then realized this whole "WebCore is messy" seems to be unsubstantiated. Somebody find some proof!
I think one of the strengths of Open Source is that developers are not under economic presure to deliver it yesterday. They have usually taken the approach of getting it right. I think this means products are sometimes longer to market, but its a trade off.
I'd like to point out here that the only reason Open Source development usually takes longer is that, as with KDE, it's largely a volunteer effort. Pay those same high-quality, principles-first developers a full time salary, and I'll bet development is just as fast as any corner-cutting proprietary shop. Open Source needs commercialization but it needs to be done properly.
What Apple did was simply fork KHTML because they insisted upon absolute control. If they had instead hired / contracted with the original KHTML developers, none of this mess would have happened and everyone would have been better off. To blame the volunteer KHTML developers for not accepting low quality patches to their hard work is asinine. The KHTML developers had put an enormous amount of hard work into making their codebase clean. I know from my own experience working with / managing Open Source projects that low-quality patches from outsiders almost always come back to bite you in the future. I blame Apple for valuing control over quality. Timeframe was not the issue.
KDE has chosen a better infrastructure than Gnome ... But Gnome is better in terms of the software they developed.
KDE's infrastructure allows for a modern, well-integrated desktop. It is an example of object oriented principles being properly applied in real world development. Everything is richly context sensitive and there is a high degree of code modularity and re-use. To give a specific example, application file dialog widgets in KDE use the same objects as the Konqueror file manager component. Unlike in GNOME, I can manage files, open a preview pane, access network resources, etc. within the file dialog. KDE has many other similar qualities that GNOME lacks. From a real-world usage perspective, KDE is highly superior to GNOME and it's no surprise why most people running Linux today use it. I would have to say the same for KDE applications -- they're generally much richer because they re-use the superior core KDE components. Examples: I know of no equal equivalents to K3B or Amarok as GNOME/Gtk apps. I also find KMail far superior to Evolution and Kopete superior to Gaim.
KDE tries to look like Windows too much for my taste. Gnome is inspired on the same ideas of Mac OS, something more elegant.
KDE doesn't really try to look like anything - it's a product of what seems to work and what users are asking for. I certainly don't see much similarity to the look/feel of Windows. (Yes, I still run into Windows a bit on the job so I'm still quite familiar..) GNOME is (now) designed upon a set of HCI guidelines developed by a handful of folks who sat illiterate users down at computers, asked them to perform simple tasks, and then decided that this must be the way interfaces should be designed. Obviously everyone else (aka. real users) don't know what they're talking about. The result is agonizingly simple, inflexible, and feature-bare software. That's not to say there isn't some great GNOME/Gtk software but the desktop environment itself is the pits. It's not MacOS X either.. it's more like Mac OS 7 or 8 in feature set / richness. Hardly "elegant" by any standards.
Every single business would go for the $80,000 propriety package. Firstly, businesses buy from other businesses, not from college students programming in their spare time, none of who want to have responsibility and their ass on the line.
Obviously you didn't even read my post because I said professional, commercial development not college students. But yeah, I agree: college students and hobby hackers are usually useless to produce quality code.
Secondly, no business cares that 12 people 'could' pool their resources together a year from now. They care about what WILL happen, and put that into a license and buy from a company whose whole job it is to support the product.
Wrong. Around 70% of all software is still developed in house. Collaboration between companies with similar needs makes perfect sense. And again, you are making silly assumptions based on taking my post at face value without actually reading. I never said anything about "x number of people" pooling resources. Get the stupid hobby hacker Open Source myth out of your head.
Thirdly, businesses expect tech support and future existence.
Which collaborative open source projects guarantee whereas proprietary software does not.. (And yes, as an IT admin, I've typically gotten much better support from professional OSS projects than from most proprietary software companies.)
Fourth, a company with employees sure shows more dedication to the product than a sourceforge archive.
Back to the stupid hobby hacker Open Source myth..
Finally, no business does things for the lower cost, they do things for the best impact for the business at that time.
Good businesses think long term as well. In the long term, a transition to Open Source solutions makes sense more sense. If a solution does not currently exist, there is often a good business case for investing in its production and deployment over a given time period.
I think you're confusing capitalism, with the the idea that govenment agencies are required to take the contractor with the lowest bid. Businesses do not operate on such a stupid premise.
Cost savings is not the sole reason many businesses are exploring / pursuing OSS. It's just an added benefit that comes with the territory. The largest benefits of OSS are control and flexibility. Of course, when you're talking $40k vs $80k, you're talking about a small company - so cost itself may be a much larger factor.
At the end of the day, people don't pay for what they value. They pay for what they have to pay for.
Exactly.. and with OSS, what people sometimes have to pay for is the labor to initially create it. That may come indirectly through support contracts or directly through development contracts. At the end of the day, you go with what does the job for the lowest cost. If a proprietary package will cost your business $80,000 to license but it will only cost $40,000 to contract required development of a mature OSS project, which is a better deal? And what happens when dozens of similar businesses are able to pool resources to make features happen? (This is why I have always advocated that most OSS should be professionally and commercially backed.. it allows for ad-hoc business alliances and incredible economies of scale)
I'd rather see this fixed in 1.0.4 or ASAP. It can't be that difficult to implement.. it's just a simple update manager and it only needs to work in Windows anyhow. Linux distros have their own update mechanisms. (although a popup reminder might be good.. "an update is avail.. [howto message customized by the distro here]")
While "many eyes" will eventually help Firefox to become almost perfectly secure, it is still a very young piece of software. The Moz. team really should have anticipated future problems and built a proper update manager *before* releasing it unto the general public. Being less buggy than IE is half the marketing drive. This is not a reputation that can stand the risk of being tarnished.
So, if anyone on the Moz team is listening, here's my vote to fix the update manager immediately instead of waiting 6mo+ for version 1.1. I don't even use Windows anymore, but I want to see my Windows-using friends continue to use Firefox.
What they really need to do is stop trying to emulate Microsoft Office. You'll never make the MS Office killer by making MS Office.
That's correct. But what we need is not another new approach to standard word processing. We need to look past word processing altogether. When you really think about it, it's a totally outdated paradigm in the modern, web-connected world. Knowledge workers shouldn't be doing their own layout and formatting unless it's actually DTP. It's time for web-based document production. Forget WYSIWYG.. that's for the layout people. The vast majority of document production can be handled by lightweight formatting via HTMLAREA style web forms. All the final formatting can then be handled by a template engine. (such as to comply perfectly with company guidelines, etc.) Document workflow / approval is all handled in a uniform fashion through the same intranet interface. Imagine the potential for automation: you update the company newsletter, somebody else proofreads it, your boss ok's it, and then it automatically gets processed into emails, letter head mailings, and the public website. No swapping around documents via email or shared folders. No manual preparation of each output form. This is the Windows/Office killer, not OO.org. (Oh yeah.. and since it's web based, people can work from anywhere.. IT at its finest)
Here's my take on it: Microsoft Research was founded not as a way of performing valuable research for Microsoft, but as a way of preventing smart people from doing valuable research. By locking up the industry brainiacs, Microsoft can virtually guarantee that no one will come up with the technology to challenge them.
Or they're just astroturfing all possible future paths of innovation with patents. I think that's a more logical conclusion given the circumstances.
Well M$ did spend good money on research (billions of dollars) and lending technology so that they can get some of that back does not seem so bad.
But then you ask.. just WHY are they spending so much money on this "research" when it is plainly obvious from the history of the industry that most software related innovation happens automatically and incrementally. Let me give you a hint: MS is one of the few players that really truly supports software patents.. and it's because they're one of the only ones that benefits.
So no, MS should not be allowed to profit on the vast majority of the "research" coming out of their labs. It's not because good ideas aren't being generated. It's because those same good ideas would have come out naturally without one company spending billions to force them out before their time. Do you see what they're trying to do? In the face of competition from Open Source and (moreso) an industry quickly shifting to vendor-neutral standards, they're trying to monopolize innovation itself. Why do you think they suddenly set up all those cheap overseas research labs? I don't see any dramatic improvement in their products. No, they're basically astroturfing all possible future directions of innovation with patents in attempt to make it impossible to compete without them still getting a piece of the action. That's not capitalism. It's unethical business practice.
Software patents are wrong and this is a perfect example of why. Hopefully most of the rest of the world will maintain a better grasp on sanity.
However, it's pretty damned good for the times where you want a machine to parse your config files and present them in a nice GUI format.
/proc and /sys virtual filesystems) However, in addition to this, store metadata about the config hierarchy in a standardized XML schema. This way, you can quickly and easily read/edit by hand, software doesn't have to spend time parsing XML on load, the security is more granular, and yet GUI tools can parse the config in a standard fashion, gaining all the benefits of XML driven configuration.
I think there's an even better solution than either straight XML or custom config schemas: a compromise that uses the best of both worlds..
As per the Elektra (http://elektra.sf.net/ project concept, store all configuration values in plain text key-value fashion but use the filesystem as the database. (Think of the
So I think what Apple did rules. They didn't ask requirements from anyone, they had a flash of inspiration, created something awesome that works good for them, and then said 'hey, if you want to use this, go ahead!' Apple never considered asking other Unices because they wanted something that worked, and also wanted to deliver too.
The problem is that most members of the "old school" Unix and even OSS community no longer have any flashes of inspiration left in them. They've been using the same crufty ideas for the last 20-30 years and as a result they can't think out of the "old school Unix" box any more. The fact that Linux and the BSD's haven't really evolved much in the last 15 years compared to OSX should be ample proof. I think it comes down to this: we need more fresh blood.. developers who are willing to look at the same old problems without a preconceived notion of what the solutions must look like. Old school folks who refuse to open their minds to radically new ideas need to just hang their towels and pass the torch. The XFree86 -> X.org transition has been a perfect example of this. The same needs to happen with all the fundamental Unix tools. Apple's ideas are interesting but they could be made even better by an OSS community actively open to change.
There's a peculiar way that we could possibly exploit the long tail phenomenon to severely cripple the "blockbuster" business model: flood the P2P networks with fake copies of all the most popular pop songs. I'm not talking about making mp3s of white noise or fart sounds. I'm talking about legally sharing indie music renamed to look like the titles of the pop crap. (preferably only quality and similar style indie stuff..) So when average users download the latest "britany spears" or "dmb" song, they instead get exposed to something else. If it's good enough, many people will actually stop and listen to it. Then, at the end, add a brief DJ-style plug of the band and song they just heard.
Possible benefits:
- more exposure for independent artists
- less exposure for pop music rubbish
- anyone who gets sued based on the filenames can stand up to the RIAA in court knowing that there is no valid evidence against them
- it would make P2P less useful for swapping RIAA content, likely reducing the number of lawsuits and simultaneously restoring P2P to what it was originally intended to be: a channel for distribution of freely licensed content.
But it's just an idea.. not a suggestion. IANAL.
Go to any trading floor or any advanced financial institution and they will rely heavily on Excel.
In this particular usage, Excel is basically a really fancy calculator and data analysis tool. However, I've seen absolutely disasterous uses of Excel as a replacement for a proper database application. I think that's what most commentors here are referring to. No spreadsheet is the right tool for the job in this case -- whether made by MS or not. Anyone who puts together a hack-job solution using Excel/Access or OpenOffice Calc/Base instead of buying or building a proper database solution is a poor IT manager at best. I therefore find the premise of the book meritless, though maybe it's a good "way out" for those who have already started down the wrong path.
"It's called the real world.."
Much of the "real world" has been stuck in a costly technology tar pit for the last 15 years or so. It's not like some sort of success story. Thanks to people like you, who obviously don't care about progress, the vicious cycle continues.
1) You have the following data sources: Oracle, Access, CSV Files, Microsoft SQL, and IBM DBase
OK.. stop right there. If those are the data sources in use for mission critical applications at some company, then the person making the core IT decisions has failed miserably. Now, I'm not saying that there's necessarily a better option than using Excel in this case, but the only reason Excel would be used is as a last resort. Starting from scratch, you would never ever use Excel to write applications or report generators. Ideally, the primary database software should be good enough that the use of spreadsheets can be banned altogether, with the possible exception of using them as a data "scratch pad."
I think this breaks down somewhat when you consider the importance of the developers. In this particular case, the purchasing company not only got the code, but the lead guy who created and/or managed the code.
Remember that the lead guy could always leave the company and maintain the fork if things went south. And if not, well, what's to say that he would have continued working on the old code (without pay) otherwise? What I'm saying is that it's a necessary risk to take. All major OSS projects need to have commercial backing so that the core developers can work full time on them. Ideally, the core developers run the show as well. Ideally, there are no proprietary "add-on" products necessary to sustain the free project. Ideally, all resources go towards developing and supporting 100% open code. So this particular example is not "ideal." However, as a user of JasperReports, I would much rather see this happen then for it to remain just a hobby project of one man. As long as JasperReports continues to be developed at least at the previous rate, this is a clear win for the Java OSS community.
That is a very good factoid that I had not noticed when I read TFA. It definitely raises the potential for the invention in terms of practicality.
At very least, it's probably worth the cost of improving waste water treatment plants. Energy and resources are going to be used regardless, so if we can get something extra in return, why not?
It has been over a decade since the last innovative new spreadsheet.. Time for something new.
..What was the huge need for expensive MS Office again?
Indeed.. time for something new. Spreadsheets, the poor man's database, are finally beginning to die out. Good riddance! Make way for RAD tools and web database applications! Any organization that stores important data in spreadsheets has something severely wrong with their IT department.
First: Will students really need to know OS solutions in the marketplace, or do we see the trend as continuing with windows/office in the workplace? This is a real argument. I personally feel that OS Solutions are going to be part of the marketplace in future years and that the time is now for Education to start to offer OS solution classes...not replace Enterprise solutions - but offer them side by side.
:) You can be certain there's a market today. Probably an eighth of the textbooks I used in college were co-authored by local profs and some weren't even that widely published. As for "teaching Linux" you probably want to avoid getting bogged down with details, like anything sysadmin related, since these are so subject to change. But familiarization with popular OSS tools and desktops can't hurt. (KDE, OO.org, Gimp, etc.) And don't bother trying to teach business types any CLI stuff. They're usually not brainy enough to handle it and it just gives them a bad impression of Linux since they don't know any better. (ie. "Oh.. this is like that old DOS crap")
The inherent problem with any classes like these is that the technology changes so rapidly and the tools actually used in the real world are so diverse. Sure, everybody today needs to know how to use a basic word processor and spreadsheet. But beyond that, where do you even start? I see business students trying to learn Access and FrontPage. Will they ever use them? Hopefully not!! I think a more appropriate approach to technology education is: "will students be able to properly handle whatever the marketplace throws at them?"
The second question, and the most important, is rather or not we can find the teaching materials (text books, etc) for the classes and rather or not the classes can be accredited. All the OS solutions in the world mean nothing to education standards if we can't accredit them. That is what I am trying to do...find text books and all the infornmation needed to accredit courses that would teach OO and Linux (as well as other basic Open Source Solutions). So far, I have not had a lot of luck in that department.
Why not get together with some profs and start writing your own textbooks? No harm making a little extra cash while scratching your itch.
I'm sorry, but the only people who really NEED to know how to use a computer are those of us who keep them working.
You know the saying, "Give a man a fish and he'll eat for a day. Teach a man how to fish and he'll eat for a lifetime." Well, teaching people how to just barely get by with using Windows, Word, etc. is an example of giving them a fish. Sure, it's the most convenient option for you, but it doesn't really help them in the long run. Change the technology and you'll have to completely re-train them since they never really understood what they were doing to begin with. (And I doubt you are closed minded enough to think that technology never changes.)
And how do the said businesses know that OpenOffice will be around in 5-6 years? Will they be able to get support or training for it? Will there be up-to-date documentation? Maybe, maybe not.
There you go repeating the standard senseless FUD again. Yeah.. you know, because large, popular Open Source projects just randomly decide to disappear one day. And you know how awesome the free support and training is that you get from MS is when you license their products. These arguments are all bunk. All business decisions involve risk and there's very little risk in going with OpenOffice for those whom it meets the needs.
Will Microsoft be around in 5-6 years? A most definite yes.
Will MS Office still be popular in another 5-6 years? A most definite maybe not.
It would take Armageddon to do in a flagship MS product line. You must be having delusions of OpenOffice granduer to think that.
Dozens of "invincible" companies and products of the past have disappeared as markets have changed. Do I think OpenOffice is a long term solution? Nope. (but neither is MS Office) However, OO.org is a viable intermediary until next-gen web-based technologies are more readily available to replace it as well.
I'm sorry, but in general most open source projects that are supposed to be replacements for commercial software just don't cut the mustard.
Open Source is only now reaching the mainstream. It is quickly emerging from its non-commercial, hobby-hacker past. Don't expect the future to be anything close to what we've seen in the past.
They don't look as good, they don't feel as good, the support isn't there, there's no guarrantee of reliability, and just doesn't look good to a business and most individuals.
Look and feel are almost entirely relative. Anything unfamiliar doesn't "look and feel" as good as what you're familiar with. I could say the exact same thing about Win/Office. I hate the basic Windows GUI compared to the much richer KDE interface. "Support isn't there" is simply a lie. And there's no guarantee of reliability in most commercial software either. Try reading some EULA's..
the vast majority of Businesses use what everyone else is using, and that's Windows and Office. Scream, cry, kick all you want u can't fight reality. Kids should be learning what the business world is using, putting on your resume "Redhat and Open Office experience" doesn't look that great to the average business running Windoze & Office.
First off, this has absolutely nothing to do with Linux. Anyone who can use Windows can use KDE. I've seen dozens of examples of this -- especially in education. So the OS is irrelevant as it is an implementation detail. It's the apps that matter.
As for OpenOffice, anyone who has trained on it can honestly put both MS Office and OpenOffice on their resume if that's such a big deal right now to clueless employers. There's virtually no difference to the average user and nearly all business users are such. The exception is specially trained secretaries who have been sent through dozens of expensive seminars on fine details of various software. This is an extreme rarity.
And BTW, do you know why businesses will not be running Win/Office in the future? It's not because Linux/OO.org will become perfect clones. (although this is an intermediate step for some) It's because the word processing and personal desktop paradigms themselves have seen their day. Businesses that stick with the old paradigm technologies will be left in the dust by those who innovate in the area of workflow efficiency. We're on the verge of a massive upheaval of the status quo and the ushering in of a new generation of business computing technology. It's been 20 years since that last happened, thanks to the effect of proprietary monopolies on true innovation.
I've heard that idiotic argument used before. The fallacy is that kids should be taught how to use specific applications that they'll find "in the real world." Wrong! They should be taught about how to use and truly understand computer technology. By the time kids in high school now are graduating from college and getting their first real jobs, MS Office may well be a thing of the past. Honestly.. do you think in 5-8 years that OpenOffice.org will not be just a tad more attractive to businesses? (or some other project if not OO.org) Or how about modern web-based document management / production systems that eschew the silly, outdated "word processing" concept that keeps today's businesses tied to inefficient workflows and excessive paper waste.
And before you say, "Yeah, but what will they need to use in college?" consider what you used in college. Was there anything that OO.org in its current imperfect state could not handle perfectly well? Typing essays and reports? Including a simple table or chart of your chem lab results?
The problem with most schools is that they focus all their energy and resources in providing students with the "best" facilities, equipment, etc. and then miss the whole point of properly educating with an eye on the future.
Supporting all of them with all the various potential support problems with varying library support would be a nightmare, even with _only_ 4 different linux distributions.
:)
In reality, most binary software runs without a hitch on most distros. The handful of support issues that do arise is no worse than the multitude of Windows-related support issues companies currently deal with. The whole situation is far far exaggerated.
Large businesses want to run Oracle for instance.
Businesses with requirements complex enough to actually need Oracle (ie. not just buy for the name) are increasingly rare. Any that do are going to run it on officially supported RedHat or SuSE as a standalone server. In other words, the choice of distro is entirely secondary to the choice to use Oracle. I'll leave aside my commentary on how high-end databases are usually a band-aid for poorly written applications / middleware...
Open source software is no more going to kill off closed source software than the internet is going to kill off printed media.
Don't bet on it. In another 10-15 years, the only "closed source" software will be application service providers -- and even these will be built using OSS. It makes too much sense and the momentum shift has already begun.
And yes, the internet will kill off printed media.. because it'll be downloaded to electronic paper instead.
I have but one comment:
Hello.. my.. name.. is.. Werner Brandes.. my.. voice.. is.. my.. passport.. verify.. me?
Because not everyone wants to be an open source software company. If linux is ever going to be used by a business, a regular end user, etc it has to be able to support closed-source programs. That means binary compatibility so a software maker doesn't have to support 15 different compiles of the same piece of software for each distribution.
First, any business running Linux is reasonably going to be using only one of 3 distros: Fedora/RedHat, Debian, or SuSE. All of the rest are special purpose distros which can be safely ignored. If you're running a proprietary shop, that extra hour of compiling/packaging isn't going to make or break your operation. Second, if you really want cross-platform software, you should be using Java and/or developing web applications to begin with. Binary-compatibility-driven desktop platforms are on their way out. Businesses are increasingly looking for ways to disconnect their IT needs from vendor and technology lock-in. Third, the availability of closed-source software for Linux is not a requirement for its success in business or even for home end users. The gaps remaining (special purpose software) may just as likely be filled by Open Source software as proprietary. It all depends on how well Open Source communities leverage the business world and vice-versa. There are plenty of level-headed (non dot-com) business strategies that have yet to be explored by Open Source entrepreneurs.