The Internet spread power around and provided for lots of innovation or at least trials of new ideas. It really seems like there is now a re-consolidation of power to servers like Google an others. This is, IMO, very *BAD* for FOSS and innovation in general. When even Universities don't have IT centers and services but merely consume services as commodities allot of the wind is going to go out of the FOSS sails. Students will certainly have fewer chances to be involved in real IT (and internships at corporations tend to be pretty lame).
I'm continuously surprised at the pro-Google [almost fan-club] attitude of so many.
The online forum idea is a good one although it doesn't necessarily have to be restricted to education.
What's the point? Students can go to google groups or a myriad other services to get the same thing. That doesn't differentiate your local service in any meaningful way.
I have exact the same story to tell as the person who asked the question. I have worked with software development, I would also want to make my own project, I have no skills in programming etc... Reading at the answers from coders has been great. I can also feel some of the pain of the OP. It is a real pain to work with software without being a developer sometimes. It would be great to have something to call your own. Vanity at play, perhaps.
Why not join an existing project? It is still "yours" once you've made a contribution. You'll get you name in lights just the same as if you create a new project - and you will be contributing [possibly] to something people actually use.
If you are *serious* about joining a project most have numerous roles for non-developers, provided those non-developers actually stick around for awhile, answer their e-mail, etc... But you've got to prove your for-real first - you'd have to do that with a new project too.
> You'll get a half-dozen "Did you try (insert > half-finished project here)?"
Which makes perfect sense and is probably the wisest course of action. By starting from an existing project you get it's name and whatever (even small) community it may have. Starting from scratch is usually pointless.
I didn't say that; I said there is little to no "low-hanging fruit". There are lots of new ideas, but they tend to be advanced/sophisticated or nuances on existing ideas/mechanisms. And thus way beyond a weeks development time.
> I have an idea for a project; it is something I would want, but googling > doesn't find me anything similar.
Really??? This seems ridiculously unlikely.
> My programming skills are not amazing, to say the least, but I can design > and QA. I'd happily learn to code, but lets face it â" getting to a good > standard would take me years, by which time I would be bored of the project.
If the project is an interest/amusement then don't even bother to start it. Successful projects are driven by *need* and a real-world problem (unless someone somewhere is funding it).
> So, my question is: in this situation, should I set up a project on > SourceForge and hope to attract some developers there? > (And if so, how do I attract developers?)
It is very unlikely that will happen. You are a tiny shrub in a vast forest. What about your project is likely to attract any lumberjacks?
> Should I try a rent-a-coder type of site and outsource the work, or perhaps > attempt to approach developers personally and share the idea, or something > else entirely?
There is no harm in trying approach developers, but I suspect you'll find many are already over committed.
> I think the project could be worth something, but I'd certainly open source > the idea if it got me the app I want. Then again, I am happy to invest some > cash in the idea, and thus cover said outsource costs â" it isn't a huge > project that I am considering, and I really think a competent developer > could probably get the thing done in a week or less (I'm not in cloud cuckoo > land here; I've worked in the software industry for over ten years, and I'm > confident that it's a fairly simple idea)
I think your cuckoo. There is a software problem a developer could solve in a week that isn't already solved? The IT industry is well past the stage of low-hanging fruit.
> To me, the question is interesting in two ways. Once I have a specific idea, > what are next steps? Then, in general, what do people do at this stage (and > this isn't specifically a software question; it would apply just as well if > I thought I had a good design for a new engine or a new type of beer)?"
Different type of products are not comparable. IT does not have the same fundamentals as the brewing industry.
And again, successful software projects [I believe] do not grow out of interesting ideas they grow out of solutions to [potentially interesting] real problems.
To scratch an itch. My company needed a fast way to convert FoxPro files to PostgreSQL, so I wrote one. Now, we're not in the database format conversion business, so this isn't something our competitors would be waiting to pounce on. Why on Earth would we want to keep it locked up? I've already gotten bug reports and feature requests that made it work better for us, so we actually came out ahead by giving it away.
Honestly, especially for projects outside a company's direct business plan, I can't think of a single reason not to subsidize FOSS. You needed it and were going to write it anyway, right?
Exactly. I contribute patches to a couple of Open Source projects. My primary employer uses those packages but they don't relate to our line of business - they are general purpose. Improvements in those packages just allow us to build on more functionality that does relate the primary line of business. Hiding them doesn't have value, it is even bad. A project with new and better features will attract more users who will potentially contribute to the project.... and we'll probably consome those features.
Unless you make software, software should not give you any more of an advantage. The product you sell and the employees you hire should give you the advantage, yes, upgrading your desktops will give you a possible productivity boost, but for most businesses technology should stay in the background.
I couldn't agree more. Bringing technology into the foreground enables innovation and [potentially] massive improvements in efficiency and methods. I go to lots of IT conferences where speakers say over and over that IT is now a commodity - they're just wrong. Through IT companies can do allot to help their competitiveness - but not if they keep technology in the background.
If I am at a restaurant, I am going to go to wherever has fast service, and cheap and decent food. If having a new cash register helps them do that fine. If having a new server helps them do that, fine. But the main object for them is to sell the most food to make the most money, that is what gives them a competitive advantage, not a new server.
This isn't a realistic example. Most businesses, and certainly that businesses that sell products or services to other businesses, do not sell-X. They sell a relationship which involves information and allot of "soft" value surrounding the product or service.
...any I worked for in the past restructured their benefits by changing employees more for their health and dental insurance and "offset" the losses by giving every employee a flat pay raise but after some calculation I found that employees with no dependents benefited from a good raise and only slightly higher insurance payments while those with families(who insured their families, at least) suffered net losses.
Uhm... doesn't this just make sense? Providing insurance to a family costs more than providing insurance to an individual. Age has nothing to do with having dependents; I know seventeen year olds with dependents.
So we either (a) socialize primarily medical service or (b) families pay more than individuals.
Personally, I'd choose (a) for economies-of-scale, efficiency, and moral reasons. But since we won't do that we have (b).
Companies aren't obligated to compensate workers who have to take on additional responsibilities to cover for the workers who chose to have children and then feel entitled.
Where I work: 600 employees, 4 IT staff (the project manager, a help-desk PC jockey guy, a web developer, and me [DBA/Sys-Admin/Net-Admin]).
We're primarily a LINUX shop (12 LINUX servers) with a couple of Windows servers and an AIX box. We are almost entirely virtualized on ESX and an EMC SAN except for our NMS (OpenNMS) and our database servers (Informix & PostgreSQL). We have roughly 300 desktops and ~50 road-warrior laptop types.
It all works pretty smoothly and while I'm on call 24/7, I rarely actually get called (maybe once or twice a year).
I feel we are adequately staffed. A well managed network is, well...., relatively easy to manage.
I'm actually in one of the rare areas that have more than one ISP. We have three available here. Our current ISP doesn't implement IPv6, so I can't use it.
Yes, you can. It is called 6to4 and makes it elementary to tunnel IPv6 over and IPv4-only leg up to the IPv6 supporting backbone.
Because it's work. Work takes time. Time is money.
If this is the attitude of your IT staff they, and possibly you, should all be fired. Technology evolves all the time, keeping it up to date is part of the job.
A certain product at a certain company (forgive my being vague, you know how these things are) has a network interface. This interface is currently IPv4 only, no IPv6 support.
So? You can run both IPv4 and IPv6 without any problems. You can route IPv4 over and IPv6 WAN, no big deal.
Because one or two products in your infrastructure don't support IPv6 is a lame reason not to roll out. None of our printers (JetDirect, etc...) support IPv6. So? The client communicate with the print server via IPv6 and the print server communicates with the "legacy" printers via IPv4. When we get new printers (which will support IPv6) it will all just work. Beats rushing to implement when the requirement suddenly arrives.
The fact of the matter is, IPv6 is a solution looking for a problem. With IP shortages and the ease of NAT/PAT,
Oh yea, that will be lovely. If anyone thinks routing and packet filtering is in rough shape now! Just wait for hopping across multple NAT lines.
Sorry, this is just STUPID! IPv6 works, it works today, on all the root DNS servers, on the entire backbone, and for many web-sites. Commercial-grade routers have been shipping with IPv6 support for years! (Cisco IOS 12.2.x). So if your equipment can't support IPv6 it has to be old as dirt or some bizarre off-branch junk. In either case it must be replaced for security related reasons anyway - dumpster time!
IPv6 is easier to route and easier to configure. All around it is just a better protocol.
Also, most of the world is using Windows XP. Can you show me where in my TCP/IP settings panel I am supposed to enter my IPv6 information? Exactly.
Do two minutes of googling; enabling IPv6 is XP is elementary.
> 1. Many consumer-grade routers do not support IPv6 out of the box.
Is this actually true? I've seen IPv6 support in numerous devices.
> 2. Some (most?) consumer ISPs do not yet support IPV6
Sadly true, although many large ISPs are using it internally. They just aren't routing it to their end users.
> 3. For both enterprises and individuals, there doesn't seem to be any cost > justification for upgrading to IPv6. > What's the benefit? It works now, right?
You don't need to "upgrade". You can run both, as we do, so you can get comfortable with IPv6. And in time you'll like IPv6 better as it solves allot of the hacks used in IPv4. It is simpler and, yes, *faster*.
I'm using flash on openSUSE 11 and pulse audio. Everything works great! openSUSE 11 has been a fantastic and trouble free experience so far (since it came out 8 - 10 hours a day five days a week).
LINUX is my only desktop and laptop OS running the GNOME desktop. I have *ZERO* problems with flash. Everything just works. I've never had a problem playing a YouTube video, my browser has never crashed on YouTube. Our corporate Intranet uses flash in many cases and always works.
Yep, this was announce a long time ago. I've even gone to.NET presentations where this was pitched, so it isn't surprising at all.
This is a way for them to automatically get the Silverlight plugin into roughly 8 million browsers. This lets Silverlight hit the ground with a large installed base.
The sad truth is that Silverlight is a really nice technology - way better than mucking about in Javascript or shoe-horning functionality into flash. Silverlight 2.0 provides a CLR on the client with the ability to download.NET assemblies.
The good news is that Silverlight works on LINUX / firefox. Microsoft has supported the Mono team and even provided multimedia codes for the proprietary media codecs. So unless you are some omg-there-are-patent-issues-with-implementing-an-ecma-standard ignorant wanker there really isn't an issue here.
> Internet company profits have zero to do with > the decline of USENET as a discussion forum.
No, it is more important that PCs don't come with news readers, and there isn't any central agent to promote/advertise a usenet group. But every stupid website can enable some horrible forum package that wastes 80% of the screen with useless gunk, whitespace, or [the worst] avatars.
> In its heyday, it was the only Internet-wide > forum.
It is still the only Internet-wide forum. Something hosted on one website by one company isn't "Internet-wide". It is there until they decide to take it down.
> It's been supplanted by web forums of every > conceivable niche.
Or mail lists. I think mail lists are what really killed Usenet. The average user has at least the software to effectively use mail lists.
> Web 2.0 beat it out, plain and simple.
Web 2.0. Pfhhht. Web 2.0 is so much white noise and nothing more. There were web-forums long before Web 2.0. They were as slow and clunky then as they are now.
We are improving what we have, it's called IPv6. Faster, lower latency, less load on routers, better address assignment, and connection-level encryption.
I don't know. Most of the conclusions make sense. (a) there is allot of interest in technologies that make obvious good sense such as virtualization and consolidation and (b) little interest in technologies that have yet to demonstrate there is any "there" there such as cloud computing. And (c) "Open Source" isn't a "technology", and of course CIOs don't care about it - they aren't paid do. That doesn't mean the same CIO won't run a department that won't use and contribute to Open Source; Open Source provides a tool-chain to provide the CIO with what he wants: the things he has to present to the board-of-directors.
As for more-or-less contractors and what not? Who knows, I don't. I still don't see the medium sized business out-sourcing, and they are the places it is much more fun to work anyway. And when taken in the aggregate employ more IT staff (or any other kind of staff for that matter) than the Fortune 1000.
Was MOSIX really about "cloud computing"? I though MOSIX was more about clusters, which isn't really the same thing.
The problem here is that most real business problems are I/O constrained not CPU constrained. So cloud/cluster/whatever is neat, but I just don't think it meets a real business need (yet) until MUCH higher bandwidth is more universally available at a MUCH more reasonable price.
Or possible these things just have no business value. I know that grid and cloud computing have no interest for me or my employer - computing power isn't the problem, it is cheap and plentiful.
Making applications with business value is the hard part, that requires smart people, which are scarce. Not just programmers, but management that can articulate what is needed, and a front-line of users who can spear head the adoption of better practices. That is the hard part. Partitioning resources, especially in a virtualized environment, isn't rocket science for most businesses.
Instead of say having an internal software dev project, and having a huge design timeframe, huge development time frame, and then 3 months for test/fix/ship, the project should be built incrementally, using the same techniques as a lot of web 2.0 startups use.
None of this has to do with Web 2.0 (or Web 3.0, or Web 4.0) or how start-ups operate (the vast majority of these start-ups will fail). This is basically just agile development practices and have been around forever.
Why does every positive or dynamic process suddenly have to be attached to whatever "Web 2.0" is?
Slashdot Mirror
The Internet spread power around and provided for lots of innovation or at least trials of new ideas. It really seems like there is now a re-consolidation of power to servers like Google an others. This is, IMO, very *BAD* for FOSS and innovation in general. When even Universities don't have IT centers and services but merely consume services as commodities allot of the wind is going to go out of the FOSS sails. Students will certainly have fewer chances to be involved in real IT (and internships at corporations tend to be pretty lame).
I'm continuously surprised at the pro-Google [almost fan-club] attitude of so many.
The online forum idea is a good one although it doesn't necessarily have to be restricted to education.
What's the point? Students can go to google groups or a myriad other services to get the same thing. That doesn't differentiate your local service in any meaningful way.
I have exact the same story to tell as the person who asked the question. I have worked with software development, I would also want to make my own project, I have no skills in programming etc... Reading at the answers from coders has been great. I can also feel some of the pain of the OP. It is a real pain to work with software without being a developer sometimes. It would be great to have something to call your own. Vanity at play, perhaps.
Why not join an existing project? It is still "yours" once you've made a contribution. You'll get you name in lights just the same as if you create a new project - and you will be contributing [possibly] to something people actually use.
If you are *serious* about joining a project most have numerous roles for non-developers, provided those non-developers actually stick around for awhile, answer their e-mail, etc... But you've got to prove your for-real first - you'd have to do that with a new project too.
> You'll get a half-dozen "Did you try (insert
> half-finished project here)?"
Which makes perfect sense and is probably the wisest course of action. By starting from an existing project you get it's name and whatever (even small) community it may have. Starting from scratch is usually pointless.
I didn't say that; I said there is little to no "low-hanging fruit". There are lots of new ideas, but they tend to be advanced/sophisticated or nuances on existing ideas/mechanisms. And thus way beyond a weeks development time.
> I have an idea for a project; it is something I would want, but googling
> doesn't find me anything similar.
Really??? This seems ridiculously unlikely.
> My programming skills are not amazing, to say the least, but I can design
> and QA. I'd happily learn to code, but lets face it â" getting to a good
> standard would take me years, by which time I would be bored of the project.
If the project is an interest/amusement then don't even bother to start it. Successful projects are driven by *need* and a real-world problem (unless someone somewhere is funding it).
> So, my question is: in this situation, should I set up a project on
> SourceForge and hope to attract some developers there?
> (And if so, how do I attract developers?)
It is very unlikely that will happen. You are a tiny shrub in a vast forest. What about your project is likely to attract any lumberjacks?
> Should I try a rent-a-coder type of site and outsource the work, or perhaps
> attempt to approach developers personally and share the idea, or something
> else entirely?
There is no harm in trying approach developers, but I suspect you'll find many are already over committed.
> I think the project could be worth something, but I'd certainly open source
> the idea if it got me the app I want. Then again, I am happy to invest some
> cash in the idea, and thus cover said outsource costs â" it isn't a huge
> project that I am considering, and I really think a competent developer
> could probably get the thing done in a week or less (I'm not in cloud cuckoo
> land here; I've worked in the software industry for over ten years, and I'm
> confident that it's a fairly simple idea)
I think your cuckoo. There is a software problem a developer could solve in a week that isn't already solved? The IT industry is well past the stage of low-hanging fruit.
> To me, the question is interesting in two ways. Once I have a specific idea,
> what are next steps? Then, in general, what do people do at this stage (and
> this isn't specifically a software question; it would apply just as well if
> I thought I had a good design for a new engine or a new type of beer)?"
Different type of products are not comparable. IT does not have the same fundamentals as the brewing industry.
And again, successful software projects [I believe] do not grow out of interesting ideas they grow out of solutions to [potentially interesting] real problems.
To scratch an itch. My company needed a fast way to convert FoxPro files to PostgreSQL, so I wrote one. Now, we're not in the database format conversion business, so this isn't something our competitors would be waiting to pounce on. Why on Earth would we want to keep it locked up? I've already gotten bug reports and feature requests that made it work better for us, so we actually came out ahead by giving it away.
Honestly, especially for projects outside a company's direct business plan, I can't think of a single reason not to subsidize FOSS. You needed it and were going to write it anyway, right?
Exactly. I contribute patches to a couple of Open Source projects. My primary employer uses those packages but they don't relate to our line of business - they are general purpose. Improvements in those packages just allow us to build on more functionality that does relate the primary line of business. Hiding them doesn't have value, it is even bad. A project with new and better features will attract more users who will potentially contribute to the project.... and we'll probably consome those features.
Unless you make software, software should not give you any more of an advantage. The product you sell and the employees you hire should give you the advantage, yes, upgrading your desktops will give you a possible productivity boost, but for most businesses technology should stay in the background.
I couldn't agree more. Bringing technology into the foreground enables innovation and [potentially] massive improvements in efficiency and methods. I go to lots of IT conferences where speakers say over and over that IT is now a commodity - they're just wrong. Through IT companies can do allot to help their competitiveness - but not if they keep technology in the background.
If I am at a restaurant, I am going to go to wherever has fast service, and cheap and decent food. If having a new cash register helps them do that fine. If having a new server helps them do that, fine. But the main object for them is to sell the most food to make the most money, that is what gives them a competitive advantage, not a new server.
This isn't a realistic example. Most businesses, and certainly that businesses that sell products or services to other businesses, do not sell-X. They sell a relationship which involves information and allot of "soft" value surrounding the product or service.
...any I worked for in the past restructured their benefits by changing employees more for their health and dental insurance and "offset" the losses by giving every employee a flat pay raise but after some calculation I found that employees with no dependents benefited from a good raise and only slightly higher insurance payments while those with families(who insured their families, at least) suffered net losses.
Uhm... doesn't this just make sense? Providing insurance to a family costs more than providing insurance to an individual. Age has nothing to do with having dependents; I know seventeen year olds with dependents.
So we either (a) socialize primarily medical service or (b) families pay more than individuals.
Personally, I'd choose (a) for economies-of-scale, efficiency, and moral reasons. But since we won't do that we have (b).
Companies aren't obligated to compensate workers who have to take on additional responsibilities to cover for the workers who chose to have children and then feel entitled.
Thank you! I think this almost every single day.
Where I work: 600 employees, 4 IT staff (the project manager, a help-desk PC jockey guy, a web developer, and me [DBA/Sys-Admin/Net-Admin]).
We're primarily a LINUX shop (12 LINUX servers) with a couple of Windows servers and an AIX box. We are almost entirely virtualized on ESX and an EMC SAN except for our NMS (OpenNMS) and our database servers (Informix & PostgreSQL). We have roughly 300 desktops and ~50 road-warrior laptop types.
It all works pretty smoothly and while I'm on call 24/7, I rarely actually get called (maybe once or twice a year).
I feel we are adequately staffed. A well managed network is, well...., relatively easy to manage.
I'm actually in one of the rare areas that have more than one ISP. We have three available here. Our current ISP doesn't implement IPv6, so I can't use it.
Yes, you can. It is called 6to4 and makes it elementary to tunnel IPv6 over and IPv4-only leg up to the IPv6 supporting backbone.
What's the downside to being ready?
Because it's work. Work takes time. Time is money.
If this is the attitude of your IT staff they, and possibly you, should all be fired. Technology evolves all the time, keeping it up to date is part of the job.
A certain product at a certain company (forgive my being vague, you know how these things are) has a network interface. This interface is currently IPv4 only, no IPv6 support.
So? You can run both IPv4 and IPv6 without any problems. You can route IPv4 over and IPv6 WAN, no big deal.
Because one or two products in your infrastructure don't support IPv6 is a lame reason not to roll out. None of our printers (JetDirect, etc...) support IPv6. So? The client communicate with the print server via IPv6 and the print server communicates with the "legacy" printers via IPv4. When we get new printers (which will support IPv6) it will all just work. Beats rushing to implement when the requirement suddenly arrives.
The fact of the matter is, IPv6 is a solution looking for a problem. With IP shortages and the ease of NAT/PAT,
Oh yea, that will be lovely. If anyone thinks routing and packet filtering is in rough shape now! Just wait for hopping across multple NAT lines.
Sorry, this is just STUPID! IPv6 works, it works today, on all the root DNS servers, on the entire backbone, and for many web-sites. Commercial-grade routers have been shipping with IPv6 support for years! (Cisco IOS 12.2.x). So if your equipment can't support IPv6 it has to be old as dirt or some bizarre off-branch junk. In either case it must be replaced for security related reasons anyway - dumpster time!
IPv6 is easier to route and easier to configure. All around it is just a better protocol.
Also, most of the world is using Windows XP. Can you show me where in my TCP/IP settings panel I am supposed to enter my IPv6 information? Exactly.
Do two minutes of googling; enabling IPv6 is XP is elementary.
> 1. Many consumer-grade routers do not support IPv6 out of the box.
Is this actually true? I've seen IPv6 support in numerous devices.
> 2. Some (most?) consumer ISPs do not yet support IPV6
Sadly true, although many large ISPs are using it internally. They just aren't routing it to their end users.
> 3. For both enterprises and individuals, there doesn't seem to be any cost
> justification for upgrading to IPv6.
> What's the benefit? It works now, right?
You don't need to "upgrade". You can run both, as we do, so you can get comfortable with IPv6. And in time you'll like IPv6 better as it solves allot of the hacks used in IPv4. It is simpler and, yes, *faster*.
I'm using flash on openSUSE 11 and pulse audio. Everything works great! openSUSE 11 has been a fantastic and trouble free experience so far (since it came out 8 - 10 hours a day five days a week).
LINUX is my only desktop and laptop OS running the GNOME desktop. I have *ZERO* problems with flash. Everything just works. I've never had a problem playing a YouTube video, my browser has never crashed on YouTube. Our corporate Intranet uses flash in many cases and always works.
Yep, this was announce a long time ago. I've even gone to .NET presentations where this was pitched, so it isn't surprising at all.
This is a way for them to automatically get the Silverlight plugin into roughly 8 million browsers. This lets Silverlight hit the ground with a large installed base.
The sad truth is that Silverlight is a really nice technology - way better than mucking about in Javascript or shoe-horning functionality into flash. Silverlight 2.0 provides a CLR on the client with the ability to download .NET assemblies.
The good news is that Silverlight works on LINUX / firefox. Microsoft has supported the Mono team and even provided multimedia codes for the proprietary media codecs. So unless you are some omg-there-are-patent-issues-with-implementing-an-ecma-standard ignorant wanker there really isn't an issue here.
> Internet company profits have zero to do with
> the decline of USENET as a discussion forum.
No, it is more important that PCs don't come with news readers, and there isn't any central agent to promote/advertise a usenet group. But every stupid website can enable some horrible forum package that wastes 80% of the screen with useless gunk, whitespace, or [the worst] avatars.
> In its heyday, it was the only Internet-wide
> forum.
It is still the only Internet-wide forum. Something hosted on one website by one company isn't "Internet-wide". It is there until they decide to take it down.
> It's been supplanted by web forums of every
> conceivable niche.
Or mail lists. I think mail lists are what really killed Usenet. The average user has at least the software to effectively use mail lists.
> Web 2.0 beat it out, plain and simple.
Web 2.0. Pfhhht. Web 2.0 is so much white noise and nothing more. There were web-forums long before Web 2.0. They were as slow and clunky then as they are now.
We are improving what we have, it's called IPv6. Faster, lower latency, less load on routers, better address assignment, and connection-level encryption.
I don't know. Most of the conclusions make sense. (a) there is allot of interest in technologies that make obvious good sense such as virtualization and consolidation and (b) little interest in technologies that have yet to demonstrate there is any "there" there such as cloud computing. And (c) "Open Source" isn't a "technology", and of course CIOs don't care about it - they aren't paid do. That doesn't mean the same CIO won't run a department that won't use and contribute to Open Source; Open Source provides a tool-chain to provide the CIO with what he wants: the things he has to present to the board-of-directors.
As for more-or-less contractors and what not? Who knows, I don't. I still don't see the medium sized business out-sourcing, and they are the places it is much more fun to work anyway. And when taken in the aggregate employ more IT staff (or any other kind of staff for that matter) than the Fortune 1000.
Was MOSIX really about "cloud computing"? I though MOSIX was more about clusters, which isn't really the same thing.
The problem here is that most real business problems are I/O constrained not CPU constrained. So cloud/cluster/whatever is neat, but I just don't think it meets a real business need (yet) until MUCH higher bandwidth is more universally available at a MUCH more reasonable price.
Or possible these things just have no business value. I know that grid and cloud computing have no interest for me or my employer - computing power isn't the problem, it is cheap and plentiful.
Making applications with business value is the hard part, that requires smart people, which are scarce. Not just programmers, but management that can articulate what is needed, and a front-line of users who can spear head the adoption of better practices. That is the hard part. Partitioning resources, especially in a virtualized environment, isn't rocket science for most businesses.
Maybe these priorities are right on the money.
Instead of say having an internal software dev project, and having a huge design timeframe, huge development time frame, and then 3 months for test/fix/ship, the project should be built incrementally, using the same techniques as a lot of web 2.0 startups use.
None of this has to do with Web 2.0 (or Web 3.0, or Web 4.0) or how start-ups operate (the vast majority of these start-ups will fail). This is basically just agile development practices and have been around forever.
Why does every positive or dynamic process suddenly have to be attached to whatever "Web 2.0" is?