You know, this is a very real but very unfortunate aspect of Sun over the last couple of years. They could have embrased Linux, and looked like they would for a while.
If they had wanted to give up all the vast technological advantages Solaris has over Linux in order to impress some spoiled teenagers, they could have. Since that isn't a viable strategy, they had to play both sides. Which tends to make the aforementioned teenagers unhappy, although many of their actual paying customers seem happy to go along.
I think you're wrong - I think the original poster was just voicing his frustration with Sun.
There is a vast difference between wishing ill on someone and being frustrated with them. The original poster was very clearly wishing ill on them.
I know that this seems like a meaningless distinction to some. It's also the reason why vast portions of the IT landscape view the Linux user community as a humungous pain in the ass, not to be dealt with, even when they could make some money serving them.
So what's left to sell? Intel boxes? AMD64 Boxes? Sparc workstations?
Do you really see Sun sustaining itself with those products? I don't.
Why not? SGI's sustaining itself with IA64 supercomputers, which is an even more limited market. Sun has much higher overhead, but on the other hand ordinary humans might buy their AMD64 stuff, in addition to the federal government.
And there are quite a few people capable of recognizing that Sun makes better stuff than Dell, even when they are selling into a commodity market.
Before you say "service" keep in mind the most expensive support plan from SUN is less then the least expensive support plan from RedHat.
The interesting thing about that is that a minimally supported Solaris machine just requires a right-to-use license, which costs a few hundred bucks on a small system. The security updates are free. A RHEL (or SuSE enterprise) license, which requires a full support contract just to get the security updates, costs thousands of dollars.
For some requirements Solaris will be a much better value.
I just hope the eds remember not to run stories from this site again.
Unlikely, since Newsforge and Slashdot have the same parent company, and it's not as if the Slashdot editorial staff are exactly dripping with integrity anyhow.
The concept of communicating directly with the guys out in the field -- the ones who were going to get their asses shot off by a tremendously superior force and knew it -- seems like a fairly well-proven idea. After all, it worked really well in Gulf War I. I have my doubts about how many Iraqi grunts had email, but we'll set those aside for now.
The truly unfortunate thing in my mind is that it apparently didn't occur to anyone to keep up this communication after the invasion when there was still a chance to win the hearts and minds of Iraqis.
We kept the communication up just fine. Our message when we rolled in was very clear: "Put down your rifles. Okay, you're fired, bye." Why this approach didn't win their hearts and minds, I don't know...
The interesting thing is that all of those messages probably gave the baath party the idea of going home (with their weapons) and waiting until the US had moved in -- thus leading, in part, to the current dilemma.
It wasn't the messages telling officers to have their soldiers sit out the battle that bred the insurgency. That message campaign was clever, and saved lives on both sides.
It was when we got there and told those guys who went along with our plan that they were all fired that we ran into problems, and drove them to some pretty desperate measures. Hopefully the reasons for that are obvious.
I'm pretty sure that Gen. Garner was replaced by Paul Bremer because of his objections to the "debaathification" methods used with the Iraqi army. That's still mostly rumor, though.
The main reason I see to move to PCI Express is that it is a fully open standard by the PCI Consortium, rather than AGP which is an Intel trade secret.
That would have been a really relevant objection to AGP about seven years ago.
It is because of this that AMD had horrible AGP support for a long time, but with the open standard that is PCI-Express everyone wins.
I don't know how long Athlon machines have been working fine with AGP, probably at least since the cache coherency problems were addressed over 3 years ago, but since that has long since been worked out, it's relevant to PCI-e vs. AGP... how?
I'm not some kind of huge AGP fan, I just hate to see such a weak argument.:)
Actually a 9800pro runs about $200 which is about what the 6600 costs. And the 9800 was not keeping up with the 6600.
I paid $130 for one recently, after a $50 "trade-up" and an online coupon on the ATI site. Sent them a crappy old card I wasn't using anymore. The nice thing about that program is that they credit your card as soon as they get your trade, rather than doing a lame rebate thing. I did have to wait a few weeks for the card, since they were backordered.
Since it's an older card and Christmas is coming, I imagine there will be more deals like this, and better deals on the non-ATI cards (Powercolor and so on). Not that I mean to strongly advocate any one card, either: you've just got to keep an open mind. There are certainly other factors, like Linux support, power consumption, video quality, TV out, bundled software and so on.
It's just that when I see anything advocating one specific card as the price/performance leader, like the intro to this story, I immediately think, "for whom?" Everyone has different needs.
In a worst-case scenario, if you manage to create an app which attracts the attention of an entity which would threaten you, at the same time you'll probably also be approached by entities that are interested in acquiring your technology (and willing to help defend it).
I would think the worst-case scenario would be having the interesting parts of your technology stolen by those entities, as terms in a legal settlement to keep them from suing you into financial oblivion. Happens fairly often, no?
I've found that Gnumeric's dependencies make it pretty much impossible to keep Gnumeric fully up to date. I imagine this isn't a concern if you're using debian-unstable or something, and constantly updating to the newest version of GNOME.
The dependency that sticks in my mind while I was trying to keep Gnumeric up to date on SuSE is "libgsf", but that might be because that's where the trouble always started. I probably would have had more problems if I'd managed to resolve that. Is there some way to make static binaries or something with the newest libraries, without removing the older libraries and breaking all sorts of annoying dependencies?
You gotta wonder why having the newest version of stuff like libgsf, down the the hundredth of a version number, is so important to each and every version of Gnumeric.
I don't think so: the industry is not exactly a wheeling-eeling freemarket: it is instead dominated by big chains and brands that have an incentive to only deal properly with other big companies and publishers or else face the wrath.
There are a few retailers over which Activision, Vivendi Universal, or even Microsoft have little leverage. Wal-Mart comes to mind. The publishers never dictate policy to Wal-Mart, it's always the other way around. No matter who you are, you're just a vendor to Wal-Mart. The shelves of the electronics department at your local Wal-Mart, and the software which is conspicuously absent from those shelves, tell a story of which software publishers took their ball and went home, and which decided to play ball, when they learned of this little reality. Most play ball.
Even among retailers with less clout, the notion of this kind of retribution seems like fantasy to me.
Add that to the fact that Valve has no way to create, box, ship, and promote the game
If Solaris is open source, it becomes a strong Linux competitor. Small businesses can deploy it onto cheap hardware. Who are they going to pay when they need support? Sun. When they need faster hardware, who are they going to buy it from? Sun. I don't know if this will actually happen, but I suspect this is what Sun is hoping.
All true, except how does Solaris being open sourced effect that situation at all? IMO the important step for those small businesses was Sun's decision to make Opteron machines a first-class Solaris platform, and sell good Opteron machines at a competitive price. Those businesses probably don't care about source code for the OS much at all, or even whether the machine runs Linux or Solaris (except as far as price is concerned).
The subtle distinctions between their various compilers
Solaris doesn't ship with a compiler, hasn't for at least seven years. If you paid for their compiler and don't like it (sucker), use gcc.
the oddness they did to X
Yeah, including display postscript was a real bastard move. Including different window managers and KDE and GNOME is really annoying too. Why can't they just stick to CDE with no features, like the other surviving Unixes?
their refusal to replace their various shells and command line utilities like "compress" with the vastly superior open source tools like "gzip"
They include open-source tools like that with Solaris 9. The tools have always been available elsewhere. Before Linux and BSD became usable, SunOS and Solaris had the strongest open source community of anyone, since they made workstations people could actually afford.
And porting Solaris code to the non-Solaris world is often quite difficult.
That is not Sun's fault. For that matter, try porting most of the stuff you find bundled with a Linux distro to any other platform... hell, just try porting all the tools you need to build it...
Because Valve knows, and has said many times, that they cannot succeed by selling online only. Retail copies are a HUGE HUGE market. And to break into that market, they had to contract with VU.
I don't buy that. I believe the stores would probably give them preferential treatment, including prime end-cap space and whatever else they wanted, even if Walnut Creek CDROM were putting out their game. It's pretty much the most eagerly awaited PC game in history.
Those contracts, as anyone who knows who's signed one, is a proverbial deal with the devil even if the publisher hasn't contributed a single cent to development.
I don't doubt that. I'm just not clear on why they let themselves be pushed into such a contract. I guess it must have been drawn before the full extent of the first half-life's success was realized.
when Valve came out with steam and started offering their product in a mode that totally bypassed Vivendi. While it is not illegal, it is certainly a dirtbag thing to do
This is analagous to musicians telling the RIAA to get lost and releasing their music over the internet instead. I can't for a second see how this could be considered a "dirtbag thing to do".
Good analogy, since production of albums and games is sort of similar, in the way they're funded at least. A lot of times the publisher fronts the money, and assumes some risk, to get the project finished. So it's a "dirtbag thing to do" (and a contract violation, I would think) if the publisher funds the production of the game and the studio cuts them out of the sales of the game.
I mostly can't believe they didn't have all the details worked out in the contract ahead of time. I don't see how there can be such large misunderstandings. It's funny, the whole thing might have come about because it's been so fucking long since they released Half Life, and their contract was drawn before the Steam technology was even conceived.
Who put the money in to the development of HL2 for 4 years? Valve did..
I wonder about that. If VU doesn't have any money in the game, I can't figure out how this is an issue at all. Why would Valve give them any leverage at all?
The original Half Life and its spinoffs have been fantastically successful, but Valve has been keeping a top-notch game studio running for over five years. I wonder how the financing of that works out, and if that's why they still need the relationship with a big publisher.
Perhaps a company can have lower revenue and market cap and still be "bigger"? It's a pretty broad term. Perhaps it's used in the wrong context, but I don't think it's necessarily a factual error.
What criteria would you use to show EA as being bigger than Apple, then?
Aspiring game programmers write games in their spare time, graphics demos, etc. and put these things together in a portfolio to apply for a paid job as a game programmer. I know; I did this, I write code, and I hire other coders. Show me another industry where you'll work for hundreds and hundreds of hours on your own time to craft a software demo to impress a potential banking/government/oil&gas employer...
Lots of modders and amateurs do it for the love of doing it, without any desire to impress a potential employer. Some of those people are uninterested in a job in the game industry precisely because of the sweathshop conditions, and the apathy amongst some (losers) towards changing those conditions.
If game programmers see no glory in that sacrifice, why on earth did they get into video games?
Presumably because they didn't know what sacrifice was involved, or were proud enough to think that they could "take it" while others couldn't.
I was there, coding like a monkey, and it was just fine.
It was fine for you. Presumably because you don't have a family or much of a social life.
What you have to understand about this is its partly the result of there being plenty of workers who want these particular jobs so much that they are willing to be screwed over.
Um... no. That statement implies that people know what they are getting into and get into it willingly. Somehow I doubt if the interviewees are told, "by the way you'll be working 80 hour weeks a month in and won't be allowed to use your vacation time."
That's why these stories are important, and shit like Forbes rating EA a great place to work is so damaging. People need to know how bad it is for most people in the industry.
Are you trying to tell me that there are *more* "conservative" (they're really radicals) people Shrub could potentially appoint than Ashcroft? Just how scary do your wingnuts get?
Slashdot Mirror
If they had wanted to give up all the vast technological advantages Solaris has over Linux in order to impress some spoiled teenagers, they could have. Since that isn't a viable strategy, they had to play both sides. Which tends to make the aforementioned teenagers unhappy, although many of their actual paying customers seem happy to go along.
There is a vast difference between wishing ill on someone and being frustrated with them. The original poster was very clearly wishing ill on them.
I know that this seems like a meaningless distinction to some. It's also the reason why vast portions of the IT landscape view the Linux user community as a humungous pain in the ass, not to be dealt with, even when they could make some money serving them.
Why not? SGI's sustaining itself with IA64 supercomputers, which is an even more limited market. Sun has much higher overhead, but on the other hand ordinary humans might buy their AMD64 stuff, in addition to the federal government.
And there are quite a few people capable of recognizing that Sun makes better stuff than Dell, even when they are selling into a commodity market.
The interesting thing about that is that a minimally supported Solaris machine just requires a right-to-use license, which costs a few hundred bucks on a small system. The security updates are free. A RHEL (or SuSE enterprise) license, which requires a full support contract just to get the security updates, costs thousands of dollars.
For some requirements Solaris will be a much better value.
Unlikely, since Newsforge and Slashdot have the same parent company, and it's not as if the Slashdot editorial staff are exactly dripping with integrity anyhow.
We kept the communication up just fine. Our message when we rolled in was very clear: "Put down your rifles. Okay, you're fired, bye." Why this approach didn't win their hearts and minds, I don't know...
It wasn't the messages telling officers to have their soldiers sit out the battle that bred the insurgency. That message campaign was clever, and saved lives on both sides.
It was when we got there and told those guys who went along with our plan that they were all fired that we ran into problems, and drove them to some pretty desperate measures. Hopefully the reasons for that are obvious.
I'm pretty sure that Gen. Garner was replaced by Paul Bremer because of his objections to the "debaathification" methods used with the Iraqi army. That's still mostly rumor, though.
Zell Miller?
riiiiiiight
That would have been a really relevant objection to AGP about seven years ago.
I don't know how long Athlon machines have been working fine with AGP, probably at least since the cache coherency problems were addressed over 3 years ago, but since that has long since been worked out, it's relevant to PCI-e vs. AGP... how?
I'm not some kind of huge AGP fan, I just hate to see such a weak argument.
I paid $130 for one recently, after a $50 "trade-up" and an online coupon on the ATI site. Sent them a crappy old card I wasn't using anymore. The nice thing about that program is that they credit your card as soon as they get your trade, rather than doing a lame rebate thing. I did have to wait a few weeks for the card, since they were backordered.
Since it's an older card and Christmas is coming, I imagine there will be more deals like this, and better deals on the non-ATI cards (Powercolor and so on). Not that I mean to strongly advocate any one card, either: you've just got to keep an open mind. There are certainly other factors, like Linux support, power consumption, video quality, TV out, bundled software and so on.
It's just that when I see anything advocating one specific card as the price/performance leader, like the intro to this story, I immediately think, "for whom?" Everyone has different needs.
I would think the worst-case scenario would be having the interesting parts of your technology stolen by those entities, as terms in a legal settlement to keep them from suing you into financial oblivion. Happens fairly often, no?
I've found that Gnumeric's dependencies make it pretty much impossible to keep Gnumeric fully up to date. I imagine this isn't a concern if you're using debian-unstable or something, and constantly updating to the newest version of GNOME.
The dependency that sticks in my mind while I was trying to keep Gnumeric up to date on SuSE is "libgsf", but that might be because that's where the trouble always started. I probably would have had more problems if I'd managed to resolve that. Is there some way to make static binaries or something with the newest libraries, without removing the older libraries and breaking all sorts of annoying dependencies?
You gotta wonder why having the newest version of stuff like libgsf, down the the hundredth of a version number, is so important to each and every version of Gnumeric.
That's really odd, since Sun was shipping Solaris with GNOME by that point in time.
Whatever happened to Old Man Murray? That site was the best.
There are a few retailers over which Activision, Vivendi Universal, or even Microsoft have little leverage. Wal-Mart comes to mind. The publishers never dictate policy to Wal-Mart, it's always the other way around. No matter who you are, you're just a vendor to Wal-Mart. The shelves of the electronics department at your local Wal-Mart, and the software which is conspicuously absent from those shelves, tell a story of which software publishers took their ball and went home, and which decided to play ball, when they learned of this little reality. Most play ball.
Even among retailers with less clout, the notion of this kind of retribution seems like fantasy to me.
I would never claim they did.
All true, except how does Solaris being open sourced effect that situation at all? IMO the important step for those small businesses was Sun's decision to make Opteron machines a first-class Solaris platform, and sell good Opteron machines at a competitive price. Those businesses probably don't care about source code for the OS much at all, or even whether the machine runs Linux or Solaris (except as far as price is concerned).
Solaris doesn't ship with a compiler, hasn't for at least seven years. If you paid for their compiler and don't like it (sucker), use gcc.
Yeah, including display postscript was a real bastard move. Including different window managers and KDE and GNOME is really annoying too. Why can't they just stick to CDE with no features, like the other surviving Unixes?
They include open-source tools like that with Solaris 9. The tools have always been available elsewhere. Before Linux and BSD became usable, SunOS and Solaris had the strongest open source community of anyone, since they made workstations people could actually afford.
That is not Sun's fault. For that matter, try porting most of the stuff you find bundled with a Linux distro to any other platform... hell, just try porting all the tools you need to build it...
Yes, they have, in articles linked to by Slashdot, even. Whether you choose to believe them or not is your choice, of course.
Google on "opensolaris" (which isn't an actual product name, afaik), possibly in conjunction with "raymond" or "perens", and you'll find more.
I don't buy that. I believe the stores would probably give them preferential treatment, including prime end-cap space and whatever else they wanted, even if Walnut Creek CDROM were putting out their game. It's pretty much the most eagerly awaited PC game in history.
I don't doubt that. I'm just not clear on why they let themselves be pushed into such a contract. I guess it must have been drawn before the full extent of the first half-life's success was realized.
Well, I sure as hell hope they are. Don't stand too close when this bubble bursts, though: you wouldn't want to get any on you.
Good analogy, since production of albums and games is sort of similar, in the way they're funded at least. A lot of times the publisher fronts the money, and assumes some risk, to get the project finished. So it's a "dirtbag thing to do" (and a contract violation, I would think) if the publisher funds the production of the game and the studio cuts them out of the sales of the game.
I mostly can't believe they didn't have all the details worked out in the contract ahead of time. I don't see how there can be such large misunderstandings. It's funny, the whole thing might have come about because it's been so fucking long since they released Half Life, and their contract was drawn before the Steam technology was even conceived.
I wonder about that. If VU doesn't have any money in the game, I can't figure out how this is an issue at all. Why would Valve give them any leverage at all?
The original Half Life and its spinoffs have been fantastically successful, but Valve has been keeping a top-notch game studio running for over five years. I wonder how the financing of that works out, and if that's why they still need the relationship with a big publisher.
Hope they can break free soon.
What criteria would you use to show EA as being bigger than Apple, then?
Lots of modders and amateurs do it for the love of doing it, without any desire to impress a potential employer. Some of those people are uninterested in a job in the game industry precisely because of the sweathshop conditions, and the apathy amongst some (losers) towards changing those conditions.
Presumably because they didn't know what sacrifice was involved, or were proud enough to think that they could "take it" while others couldn't.
It was fine for you. Presumably because you don't have a family or much of a social life.
Um... no. That statement implies that people know what they are getting into and get into it willingly. Somehow I doubt if the interviewees are told, "by the way you'll be working 80 hour weeks a month in and won't be allowed to use your vacation time."
That's why these stories are important, and shit like Forbes rating EA a great place to work is so damaging. People need to know how bad it is for most people in the industry.
Two words:
Ed Meese