Slashdot Mirror
Available To The Right Buyer: Sun Microsystems
antediluvian writes "The Seattle Times reports Sun Microsystems shares surged forward on speculation the computer maker may be bought by a rival company. Prospective buyers could include Dell, IBM or Hewlett-Packard. Computer sales of rival companies have been outpacing sales of Sun's machines. Over the past three years Sun's stock has declined 92 percent."
...should offer to buy them. At a ridiculously low price. Turnabout, being fair play, and all. :-)
Their buyer is SCO?
NOw that would be ironic wouldn't it?
Don't Tread on OpenSource
I really, really like Sun hardware, and I'd hate to see it all go the way of the Alpha. Plus, what would happen to Java, I wonder?
PS3 client and Sun server backend for on-line gaming
Over the past three years Sun's stock has declined 92 percent
Gee, do the stock prices of three years ago mean anything? Yahoo and Amazon must also be bought!
Hey, me and my buddies poured out our loose pocket change, and dug around for some coins under the cushions on the couch.... and I think we've scraped up enough to buy Sun ourselves! The first thing I'll do is bring back the "Mr. Coffee" JavaStation, and then fire Scott McNealy. Second step is to get Ed Zander back. Then, PROFIT!!!!!
Why doesn't the open source community take over Sun? Now that would be the day.
N1 is a new IT architecture from Sun. I think it is awesome new technology/architecture, but I also think there is no market for that currently. N1 was in wrong place at the wrong time. There are lot of other things that need to be done before N1 can be implemented anywhere.
What will happen to N1 after the acquisition? IBM already has a similar product callled Tivoli. If IBM purchases Sun, N1 will either be slashed or integrated into Tivoli. Any thoughts on that?
Consensus is good, but informed dictatorship is better
It's a rumor, it sells shares of stock. Is it workable? No. However if they were to merge with say Cisco or HP that would be great. Both have their limitations. Sun is way the heck ahead in the 64 bit computer game, having an army of 64 bit gurus, a stable OS, and a very well respected CPU.
Sorry about the writing. Robot fingers, you know? Cliff Steele in DOOM PATROL #23
Too bad Apple just spent their allowance.
This could be a good move or a bad move for the computer industry. So only time will tell. Personally I would not mind if they were bought out by IBM, only because of the hate that I have for Dell.
Dell, IBM, Sun and HP have been openly slating eachother for quite some time right now. Funny cartoons around will make it very difficult for any company to be happy with their overlords^H^H^H^H^H^H^H^H^H bosses who have just bought them out. If anyone buys Sun then they will probably kill it off reather than have to manige a very angry workforce.
Mouse powered Chips, Open source Processors and Lego
Anyone have an idea what kind of steward HP would be for Java?
To be honest, I'm not sure why anyone would want Sun. Don't get me wrong, they have some great technology and are a good company. But they remind me a little of Digital pre-Compaq buyout, great technology which became irrelevant. The move is towards x86 technology, and with 64 bit x86 become more and more viable, there is simply less and less need for the premium price paid for Sun products.
Don't get me wrong, I'm not saying they'll be dead in a year, just I can't see their sales growing much, and quite possible slowly reversing. There are still some very high end applications where Sun products may well be the best product for the job, but they are painting themselves into a corner - that niche is getting smaller and smaller as x86 gets better and better.
Read reviews of shopping cart software
Can I bid for it on Ebay?
Be you Admins? nay, we are but lusers!
I should imagine IBM are after their Java division. They're probably not interested in the servers. Whether they'd just leave them as Sun, or buy the whole lot and wind the server business down over a period of years I don't know. If they do get the servers, expect to see a lot of work go into Linux on Sparc. Mark
"XML is like violence. If it doesn't solve your problem, use more." - Anonymous Coward
Over the past 3 years, lots of companies, good and bad, suffered stock point losses. But just because there's a # fixed to a company says nothign about the company.
There are lies, damn lies, and statistics. Stocks are a statistic representing the value of the company. If we put the head of a steel making company in charge of visa, visa won't lose points. Over time, prolly, just 'cause the steel head (heh) will affect the company's performance in the long rung.. prolly.
--
"I'm not bright. Big words confuse me. But Wanda loves me and that should be enough for you." - Cosmo
They've got plenty of [for mods: ill-gotten :)] money, and they've been looking to capture the server market as well.
Don't you see, it's all happening again!
A computer tech company, becoming irrelevant, trying to get bought out!
Can't you see what will happen next?
THEY WILL SUE IBM SAYING PARTS OF SOLARIS ARE IN LINUX!
(/me removes tongue from cheek now).
www.eFax.com are spammers
I dunno... does Apple really want to buy Sun?
Apple would have to capitalize on Sun's strength - the data center.
In addition, they'd have to save some serious operating $$$. To me, that means heading in the same direction in terms of OS and in terms of CPU architecture.
I'm not saying it's not doable. But doing so would mean BIG changes to the customers (either of Apple, or of Sun, or both).
And customers just don't like big change.
Both companies are leaders in terms of technology, and Jobs is pretty darn good at marketing. He is a good desktop visionary. But can he understand the datacenter?
An institutional buyer made a large purchase of Sun Stock. That fueled rumors about a buyout, but it seems a lot more likely that after reporting (admittedly very modest) profits in the last quarter and one analyst recently shifting Sun to buy, some institutional buyer wanted to get some "bargain" stock that they think will appreciate well in the coming years. Given how steady the stock price has been between 3 & 4 dollars, it does seem likely that it's bottomed out, so unless you think Sun is imminently going out of business (which I sure don't) this kind of buy seems to make sense more from that standpoint than from any bs about being bought by a bigger player.
As far as it goes, Sun's culture is so antithetical to IBM and to the "new" HP that I can't see either of them wanting to take Sun on....
7 November 2006: The day Americans realized corruption and incompetence weren't addressing 11 September 2001
Hey, maybe a company named "Cobalt" can buy them and then run them into the ground!!
I would prefer to see Sun come out with more Intel-competitive products than stick with their niche. They have the potential to make astounding lower-end hardware, and if they could keep the prices low enough it would be fantastic to have more competition with Intel's lines of chips. AMD is proving to be valuable competition, but I'd also like to see more desktop hardware that doesn't aim for compatibility with Intel.
Developers: We can use your help.
At UBS Warburg, Jack Francis, co-head of equity trading, said the sudden surge in price followed a 5-million-share block trade, considered to be a very large buy by Wall Street standards. "That was spurring stories of a potential takeover, which doesn't make any sense at all but did add fuel to the upside," said Francis. "The rumor doesn't hold a lot of weight, but in a market like this it gets people off the fence who are looking for any story that could generate alpha."
Anyway, Sun are currently valued at $12Bn, and have $5.5Bn sitting in the bank.
Sun's stock fell 92% in the past three years? Jesus.
Oh wait. Everyone's stock fell around 92% in the past three years.
[insert witty quote here]
All the anti-Sun FUD that keeps getting posted to Slashdot reminds me of the anti-Apple FUD that was all over the media a few years ago.
Speculation about IBM or HP buying Sun now is probably just as groundless as speculation about Sony or Disney (or Sun) buying Apple five years ago. Yeah, they're not doing as well as they used to, but the whole industry isn't, either.
I think Sun's main problem right now is the same problem that Apple has right now: getting hardware that customers will perceive as being equal or superior to x86 in price/performance. It looks like SPARC will get there eventually, but not soon enough; I imagine they'd either have to use Opteron/Hammer on their low-end machines, or somehow make very inexpensive 1-4 processor workstations and servers to leverage SPARC's scalability (it is, after all, the Scalable Processor ARChitecture) and Solaris's superior SMP support.
I'll admit that I have many reasons to Want To Believe that Sun will still be a strong presence in the industry when I graduate from college, but I do seriously think that rumor's of Sun's imminent death are greatly exaggerated.
Ubi dubium, ibi libertas.
I like Sun.
Well, I don't like Solaris, the hardware is overpriced and I detest Java, but it's still a good company.
Maybe it's because of the Self programming language.
graspee
Apple's merging with Sun. They'll call the new company Snapple.
Karma: Excellent (In Soviet Russia, karma pimps YOU)
For smaller server purposes, 64-bit x86 and Itanium may be a more economical choice.
But if you need a large memory bandwidth, I think probably still beats out Itanium, and definitely beats x86.
If you need a whole shitload of CPUs in one box, Sparc is also a better architecture - even if Itanium can scale up to hundreds of processors, there's no OS that runs on it which can properly handle that many.
What if MS buys Sun and then ... hehehe ... strangles/suffocates/dismembers Java?
While we're at it, why doesn't the open source write an ext2 defragmenter?
I am surprised no one mentioned EDS. HP and Dell both want very badly to become like the current IBM, who makes a ton of money on both hardware, OS, and services integeration. I think that should discount HP, Dell and IBM because the merger doesn't bring alot of new things to merged company.
EDS however was the top services company until IBM decided to go into high end consulting and services business. So... it seems an EDS / Sun merger would put them both back in IBM's league. A customer could chose IBM / zOS / db2 / mainframe for a big account or EDS / solaris / oracle / sunfire at a discout.
It also would be interesting since EDS reportedly uses big Sun servers all over the world...
just my $0.02.
Sun still makes an assload of money.
"Makes an assload of money" is not synomymous with "incurs huge losses"
One of the most misused words of the English language is "ironic." Whenver people try to sound overly sophisticated and intelligent, they tend to use that word and make their statement or argument more intelligent, and so many times I sit there thinking, "No... it's not ironic at all... maybe a coincidence, but it is not ironic." The song "Ironic" by Alanis Morissette doesn't really have anything ironic in it at all. Almost all the scenarios that she portrays aren't ironic, but are umm.... bummers. So instead of her singing, "isn't it ironic," she should sing "isn't it a bummer." Now what would be ironic is if she deliberately titled the song "ironic" knowing full well that there's no irony in there at all
I've read a lot of bitter comments on this forum about the fact that java isn't an open technology. This hasn't mattered much to me because of their community process, and otherwise open attitude, and open off-shoot projects (STL, Struts, Tomcat, etc).
I'm not trolling here at all -- I wonder what the implications for Java could be in the face of a buyout. Obviously, that would depend in some part on the buyer. And there would always be the GNU foundations free implementations. OTOH, perhaps a buyout could actually prompt Java to be handed over to a standards board.
These are rumors though, and I can't recall ever hearing a merger/buyout rumor that actually panned out (maybe I just hear bad gossip, though), so I don't put a lot of beleif into this story. It's just speculation about what Sun might do in an x86's world .
I will say that it's interesting to me to see how it's usually not the case that the best technologies survive. However, when looked at from a natural selection viewpoint, one realizes that since the computing ecology is shifted towards MS products, the x86 architecure hardware has an advantage, even though it isn't the best.
Change the OS ecology, and x86 may not be the de facto architecure...
All one has to do is hit the right keys at the right time and the instrument plays itself. - Johann Sebastian Bach
Because they already have
Marxism is the opiate of dumbasses
I've been thinking that Sun would get bought for a year or so, and I think that it will suck for computing in general. The way I see it, if Sun were to be bought, then their product line would be reduced to their larger machines just like the Proliant servers are pretty much the only thing that survived the Compaq acquisition. This will mean a drastic decrease in the number of people using Solaris, and it will be a nitche/legacy product.
Solaris is an incredibly mature OS. Just read the manpage for the sar command some day. Also there is Trusted Solaris, and F-C2 security certification, etc. Linux is my favorite OS, but Solaris definitely has my respect for its stability, scalability and maturity. And the number of users of Solaris would decrease dramatically if Sun were acquired. Think about how different the Microsoft userbase would be if they suddenly had no desktop presence and were only servers.
However, I also think that Sun should hold in there. I mean a 30% drop in sales, thats almost to be expected in todays economic situation. I mean travel is down like 50-70% in some places. Also one has to keep in mind that Sun machines have a longer lifetime on average than say a PC, so thier volume of sales will be lower in comparison.
Sun does need to get the performance of thier Sparc chips up to the others. Thier performance is a big drawback to the pricetag of a Sun machine. But everything else about thier hardware is top noche. I mean they are so anal with their hardware that they put lot numbers on each of thier ethernet cables. And their machines are just perfectly engineered. Any box that I've been inside of, I never thought "Why the hell did they put that there?".
But, who knows maybe this will be a good thing. I mean all of their employees will go to work somewhere, and maybe Solaris and NFS sources will be opened up.
However, if it were up to me, I'd just prefer Sun sticked around for a while.
I don't think this is a hardware war mate. The name of the game (IMO) is Operating Systems. Linux will run on any processor, and with the appearance of various clusters in the market, which are better than a single powerful computer anyway (redundancy, resilience, cost, just to point out a few), the hardware has become irrelevant.
In an age where the operating system does not care which architecture it runs on, it becomes a question of social classes again - the high class people will buy expensive super-powerful processors/fast ram/amazing motherboards/huge drives, while the low classes (third world countries) will buy VIA's Eden processor (x86 clone, only just reached 1Ghz speeds). And then does it really matter? They will all run Evolution/KOffice/OpenOffice for you! Just a matter of convenience... and that's just the same as with anything else... I live in a shoebox of a house, but my rich neighbours own a huge cottage.
All those moments will be lost in time, like tears in rain... time... to... die...
HP would not even bother about Sun right now because it does not want to bite off more than it can chew. Investors would not at all take kindly to the acquistion of Sun by HP. HP right now is trying to fend off the dog that is Dell. HP does have about 13.2 billion $ in hand (Biz Journals) but it will probably not want to mess with it right now.
Removing Sun from the competition would help the server market by bringing some consolidation.
What will IBM do with the Sparc chips? It is not likely to dump it for a while but after 2-3 years it may just move to Itanium and its own PowerPc chips.
Sun has already brought in x86 systems in the lower end. Both Sun and IBM are adopting AMD's Opteron for lower and mid level systems.
We have also got to remember the FTC. If IBM does bid for Sun then expect them to go through a tough scrutiny so as to avoid a monopoly status in the high end server industry.
People know that Sun is able to keep customers only by chanting the reliability and customer satisfaction song. Its Ultra Sparc's are falling behind in performance and it is probably only with the Sparc V's that it can gain any semblance of competitiveness. And when are the Sparc V's going to come out? 2005 at the earliest.( News)
Would Dell bid for Sun? Dell certainly can because it does have quite a bit of cash sitting around 9.1 billion $ as of Dec 2002 (Motley Fool and Yahoo ).
What is Sun's market capitalisation? As of March 19, it was about 10.73 billion $.
Dell does not have a foothold in the high end server market because it does not spend much of R & D as opposed to HP, IBM and Sun. Acquistion of Sun could be a easy way to compete with HP and IBM. Dell's entry could help reduce the prices of high end servers like Dell has done to the desktop market.
If this story is indeed true then it would be the most talked about merger. Competition for customers paying money for big tin has only gotten worse after the tech meltdown.
Personally I feel that the Sun bid is just a rumor like the Universal/Apple deal. If anyone is to believe it, then Sun or whoever is buying them have to publicly state that they are looking into this deal. Maybe the coming weeks will tell us more.
I bet it's IBM - they have invested a lot more in Java than Sun has over the past 2 years; and Java fits in with their old strategy of one platform running on several different levels of machine; and IBM also has a history (and the cash) of buying up expensive companies for just one aspect of them.
I guess we'll see.
These kinds of rumors are a recurring phenomenon in this industry. Check for example "When will IBM buy Sun?" which is over a year now.
Sun has a market capitalization of around $12 billion (at its current stock price of $3.75).
To buy it with a good premium would mean a huge investment.
And considering that Sun always stands alone and that its products -- hardware as well as software -- are not really compatible with the rest of the industry, anyone who would buy Sun would only buy its customers. But for how long?
Sun customers are among the most loyal ones.
And you can believe me: I was working for one of its competitors.
There's nothing wrong with Java, other than the fact that it's not as fast as C++. People just like to whine about it because they're worried they might have to learn something new one day.
This space intentionally left blank.
1) Microsoft owns the desktop 2) Microsoft introduces .net, which makes Windows the only viable server platform if you want to use .net clients.
2.5) (you can argue that any web services platform can be used for .net clients and while that might be somewhat true it is less convenient. Also, how long will it be true?)
3) step 3, Microsoft owns the server room
The Sun and Apple combination are, put together, the two only companies that can challenge that.
In the long term there is a possibility that IBM with Linux may challenge Microsoft but that day is not even close.
Well, that is my take on it.
The Internet is full. Go Away!!!
That's an interesting thought.
Two excellent computer companies in opposite markets.
Apple has the user end, Sun has the server end. They both make top notch quality, and have excellent reputations for good reliable products.
That idea could be really good, or really bad.
Who knows.
If they win the case Sun will fund its own takeover :)
crapfilter evasion goes here.
crapfilter evasion goes here.
While I hate to say HP-UX is a good OS, it is certainly an OS which runs on Itanium and supports 64 processors.
The new HP Superdome machines with Itanium2 are more powerful CPU-wise than anything Sun makes at the moment.
Ewan
Well people aren't in just a big rush to roll out .net clients.
Where did you come up with Microsoft owning the server room?
Funny you just proved your own point of not understanding irony..
Wh owns Unix IP rights? SCo
Who Set up to Make a brand New UNix(at that time)/ SUN
Now that is irony that old style unix xompany woudl be able to buy a new sytel unix company..
Of course you have no knowledge of SUn or SCO so why are you in this thread?
Don't Tread on OpenSource
Off course, the one company that has not been mentioned is Micro$oft - and they have loads of cash floating around. They can easily buy out Sun, and get rid of one of their most vocal opponents.
that Sun wasn't brought down by Billy Gates. It was brought down by Linus, E.S.R, Bob Young, Richard Stallman, IBM, and everything GNU.
+2 cents.
Now where did I put my wooden teeth... ?
Read the EFF's Fair Use FAQ
SCO has threatened to cut off IBM's licensing of Unix in the current $1B lawsuit but Sun has a perpetual license to its Unix IP. The acquisition of Sun by IBM might derail any outcome of the SCO/IBM litigation that would otherwise have resulted in serious consequences for AIX products. With Solaris available as a quick fix for any AIX licensing problems, IBM is poised to shrug it off and forge ahead claiming licensing from Sun IP covers AIX.
Furthermore, the acquisition of Sun would put IBM in the catbird seat of *nix vendors with AIX, Solaris and Linux products for practically any business niche. This acquisition, under the current circumstances, makes a lot of sense but I'd certainly hate to be in the Sun channel if it happens; IBM would almost certainly cut them off in favor of internal sales and marketing.
Expect to see a lot more ex-Sun shops turn into Linux shops if this goes through.
No one ever had to evacuate a city because the solar panels broke!
Face it. Sun sucks. Period.
I've been working with Sun Solaris and Sun SunOne products for the past two years and I have to say I've NEVER gotten worse support from ANY company (yes, even Microsoft, and their support sucks ass).
Example. The person maintaining our Enterprise 220R before me was vary lax about applying patches to the OS. So, I go in to try and clean up and find out there are neary 300 patches to apply after a year and a half with no cumulative patch at all! So, I hunker down and start reading the readmes for all the patches that need to be applied. (I got the list from their ridiculous PatchPro Java app. Just the fact that there is no progress indicator to tell you if it's crashed or still running or what is ridiculous.) I find that some of the patches require downtime. OK... I can schedule that. What else? Some of the patches say that they may actually make the machine non-bootable if incorrectly applied. ! So, I finally call Sun for support on trying to apply the patches. What do they tell me? "Do you have a test box?" Do I have a TEST box!? Fuck no! Why should I need to run a test box for every Sun server I have? To make Sun more money? NO thank you!!! So I decided to say screw this and just procede with upgrading our SunOne Messaging server even though the OS isn't patched, then I'll give the OS a shot later. When I talk to SunOne support, they say, "Do you have a text box?" !!!!!? WTF?! I tell them that I don't. They say, that I can try to do the upgrade but I would be much better off with a test box. I ask them how much downtime can I expect if I apply all the Sun Solaris OS patches and SunOne Messaging patches. They gave me a quick run down and I can expect up to 36 hours of downtime!!! Again I say, W-T F!!!!?
After some discussion with my supervisors, we determine that we wil buy a new box to install a newer version of Solaris and the SunOne Messaging server. I do this. Then I call support to find out how I move mail and users from one box to another. They give me a ton of bizarre and very byzantine approaches to do the same thing. Each tech tells me something different which invalidates what the previous tech told me. As I progress, I get to learn to hate Sun more and more and more. I know it's an alternative to M$, but goddamn!!! In the end, I start to feel that I'd be better off if would just research some open source projects and learn all the little inconveniences they bring considering that al of Sun's products feel like really bad alpha level software. So you can just surmise that I H-A-T-E Sun. Considering the amount of money they charge us for support, I'd really love to see some good service, but this is never going to happen. It just seems that Sun's main philosophy is to work with one hand tied behind their back, one eye missing and the other looking at everything through a prism and a mirror and then learn to cope with it. I for one won't be sorry to see these assclowns disappear.
Fortunately, I've convinced my bosses to NEVER buy another Sun product again. We're an OpenVMS shop with a legacy app that is being ported to Solaris and HP-UX. The company that owns that legacy app really wanted us to go the Sun route, but we chose HP-UX because we are a Digital/Compaq shop. I will be glad to see those HP-UX boxes come in. Fuck Sun.
That being said, I'm not a Java-hater. I don't think it's an awful language. My main complaint is how verbose it is. Now, compactness is not always a measure of a good programming language (consider APL, or some of the more obscure Perl regular expressions) but it's a good place to start. It irritates me to have to write ten lines of code when one line in some other language would do; hell, it even irritates me when a line of code in one language is equivalent to a line of code in the other, but the first takes five times as much typing.
But what Java did do -- and this is valuable -- was bring the idea of object-oriented, platform-independent, and VM-based languages into the mainstream. They were around before Java, of course, but none of them had anywhere near its impact. This is unreservedly a Good Thing, and Sun should be thanked for having done it, whatever happens in the future.
The correlation between ignorance of statistics and using "correlation is not causation" as an argument is close to 1.
No surprise, considering they are the dot in dot-com ;)
Getting rid of ... is like getting rid of cobol...it'll take a while
That's a big YEP. Eventually once all IT jobs are outsourced to those c*cks*ckers in India, we'll all be relics of a byegone era. They'll have a picture, in a museum somewhere, of us next to all the other extinct species.
It's little consolation to most, but at least programming started out as a hobby for me. I'm perfectly willing to let it become one again.
Agreed. Think there is gnashing of teeth now over Office licenses???? Imagine how bad Microsoft could stick .Net licensing up your backside once your whole company runs on .Net.
The cash they have on hand NOW would look like pocketchange compared to that scenerio.
God forbid the day that Microsoft controls anything more than it controls right now. It would bankrupt corporate America!
There are numerous things wrong with Java. If you can only name one(it's not as fast as C++), then you really don't know the language. Oh, and there are people who claim to have examples of Java beating C++ in performance. I have yet to see any of these that actually deliver though.
Other than all the people that have to USE those Java apps and know that it should be faster and conform to the UI specs.
See the difference?
It would complete their enterprise app technology stack (SunONE web/app servers with a big time DB) and get them into the hardware market. Might be interesting....
Other tidbits:
Sun's memory bandwidth is less than desirable, on par with a low end PIV. The Itanium has the highest sustained memory bandwidth outside of shared memory and other single purpose "super computers".
IBM never fell more than 30 percent.
Three years ago, their high was just over 100. This year they haven't been below 75. They closed the week on a steady upward climb. They're just short of 90.
Anybody wonders, why this rumour was quoted from the Seattle Times .... how far is that from Redmond???
Browsers shouldn't have a back button!! It's all about going forward...
seems a natural. They are ALL about java these days. Plus, they could see pre-configured servers.
DO NOT DISTURB THE SE
Maybe Apple would release J2DK1.4.1 for Linux/PPC if they buy Sun.
If M$ buys it, we're screwed. Java will suck complete ass within a month.
As for OS X Server on UltraSparcs, the problem of software arizes. Just like NT on Alpha, the majority of software would be built for the "original" architecture. A closed source OS can't thrive on two architectures at once.
You can't judge a book by the way it wears its hair.
IBM current stock price: about $85
IBM current stock price times 10: about $850
Unless IBM was at $850 three years ago, it hasnt even dropped 90% (BTW, IBMs all time high is around $135-$140).
Move on. There's nothing to see here.
IBM purchasing Sun would be a big win in my opinion. HP has yet to prove that they really have a handle on the software side of their company, while IBM has done more with Java and Linux than Sun ever did. Of course they might also screw Sun up even more in trying to merge it into the corporate behemoth of IBM.
what the hell?
OpenVMS, the world's most stable operating system, will be running on a cluster of itanium processors with fault-tolerance and fail safe capabilities built in.
No, that's not the only reason people whine about it. I learned C++ and Java at almost the same time (C++ about a year earlier) and I, like a lot of other programmers, can honestly say that IMO C++ is just a better language than Java. It's more compact, easier and more pleasant to write, etc.
You had me right up until "easier". There is a reason C++ journals have sections devoted to obscure sections of the standard and how code might not compile the way you would expect it to. It's because C++ is not simple. Powerul? Yes. Easy? No. Not to mention the differing implementations by different compilers. Ugh. No thanks. I'd rather spend my time working on solutions instead of fighting the language.
If you can only name one(it's not as fast as C++), then you really don't know the language.
You didn't name any either, cap'n. Beyond that, (and for the last time, Jesus) the entire concept of Java is that it is implemented on an imaginary computer, which is then emulated by a real one. While this was not an original concept by 1996, conceptually it DOESN'T EVEN MAKE SENSE that Java could be faster than a small language that compiles directly into native code. That is, however the beauty of the application of Moore's law, as we can fit more transistors on a chip, the more memory we fit on a chip and the faster our processors become.
If it takes 2 cycles to increment an integer variable in memory in C, and it takes 4 to increment an integer in memory in Java, suddenly a difference of two clock cycles doesn't appear so appalling when you're running a system with a clock speed of 5000 cycles/sec. Granted that's not today, but I seriously would not expect that as processors become faster and faster that we continue writing general purpose code in vi. Granted there will ALWAYS be a place for very efficient, lower-level languages (the K&R book does describe C as a low-level language, though obviously C++ is a further abstraction), they just may not be at the forefront of computation.
--- What
In the world of business, where are they now with those technologies? I am not speaking historicly, I am talking about the present. While x86-64 and IPF are 'just around the corner' they don't have a large amount of 64bit software resources. The free software users/developers are not the alone in their reluctance to move up. Sun as a wonderfully supported backend infrastructure, their front end sucks. The same can be said for Linux. The windows world has the opposite problem. All glitz no bang. Generally I find that the fault often lies between the keyboard and the chair.
Sorry about the writing. Robot fingers, you know? Cliff Steele in DOOM PATROL #23
I work in IT and a while back there was some talk internally that IBM would be taking over Sun. One of their products, Eclipse, is a not so subtle sign that IBM aspired to take the lead with Java. I believe that going forward the only value Sun would have is Java since hardware has improved dramatically for x86/64bit and PowerPC architectures, and the fact that nearly all of the Sun products that I have had to use are always of poor quality.
.NET will easily take over. I know that IBM will be able to produce the kind of Java technologies we know should have been built years ago, but Sun never got passed suing Microsoft to realize innovation, market share and better products are what matters, not patents and law suits.
By comparison, IBM has done a great job with producing great software and new frameworks. They have also contributed a great deal of software to the Apache/Jakarta and XML projects. They are already the leader in Java technology, Sun just owns the patents and copyrights behind it. IBM needs that to really allow Java to take off.
If you leave Sun as it is for too long, it will kill Java and
I would like to see Big Blue as the driving force behind Java.
Brennan Stehling - http://brennan.offwhite.net/blog/
--sort of one of those coinky-dinks. Last night late before I retired I had a flash that sun was going to be sold-to apple! I mean, it was almost like a daydream/vision sort of thing. It seems a logical progression to me, and follows apples concepts of an integrated hardware/software business model. They would have an *almost* complete vertical "computing" business then, from the lowliest desktop to enterprise class 'stuff' in their stable.
hmm, no more chip worries perhaps? hmmmm
Sun has some really sweet hardware that can truly scale and handle tremendous load. Builtin hardware features like headless operation even at the firmware level over a Serial port. You have to understand that a simple Netra or Sunblade is nothing compared to their highend servers. But you pay a premium for those servers. Still Linux still cannot scale as well as Sun so for Big Iron, you really need Sun 64bit SPARC systems.
/root partitions after reboot. This allows you to keep production systems online while copying the udpate to the servers. Reboot after the next business cycle, etc. Very slick.
Next, Solaris is very mature. There are a lot of features that allow you to do a remote upgrade of the OS and schedule a reboot when it's proper. i.e. it swaps the
In the software world Sun is to Linux as Oracle is to PostgreSQL. It's commercial and expensive but it still has a real reason for existing as it is able to scale to large systems better and it offers features not yet available on the open source alternative.
So, one must do an evaluation. i.e. Determine what systems meet your needs best. Linux for lowend - midrange servers and Solaris for highend servers. PostgreSQL for lowend - midrange servers and Oracle for highend servers.
Hey, why not do both? My whole complaint against the corporate world was that it would dismiss Linux out of hand and not even look at it. Bah... It's a Hacker OS... So they would rollout NT or Netware systems where Linux made more sense.
Afraid of using Linux on truly mission critical systems? Go with old Sun stuff that you probably have on hand. Then have an engineering group assemble a Linux solution and run them in parallel. Setup a failover and then test them live and surely you will be impressed.
On our Sun systems we run something like 10-15 mission critical J2EE systems using Weblogic. For a very reduced price I could assemble a 1U blade rack of dual Xeon blades and run Apache, Tomcat, JBOSS, and PostgreSQL instead of UDB/Sybase. Heck for the same price I could rollout 4 racks just be heavily redundant! (x86 hardware is not as reliable as Sun SPARC systems. We just lost a system board on an old Sun box that is 10 years old. Short of hard disk failures that's a first for us.)
I have more problems with Weblogic then I do with Sun hardware. I feel it is extremely over-priced. JBOSS can keep up for the most part and we don't even use the Weblogic features that exceed JBOSS functionality. I'd be happy to run JBOSS instead of Weblogic on the same Sun hardware! That's how much I hate Weblogic.
Really? Funny that. Banking is one of the industries that has adopted Java the most. The application server market (what transaction processing systems have evolved into today) are split about 80%/20% between Java (BEA, IBM, and others) and Microsoft. Java dominates transaction processing systems for all new development. It is also a very common choice for legacy software integration.
--Be human.
sun is a technology company. dell is a reseller. for a fortune 500 company, they have one of the lowest r&d budgets. all their r&d is done by intel, microsoft, and the OSS community. all they have perfected is the most efficient way to build a pc and ship it to you, oh yeah, and make cool ads, dude.
sun is a true tech company. so is ibm, and so is apple. you might hate/love each of them, but you can't deny they innovate. dell wouldn't know what to do with java anymore than microsoft would. of course, figuring how much gates' ass mikey dell kisses, guess we know what dell would do to java. and sparc. and solaris. and...
My problem? I was perfectly gruntled, until some numbnuts came by and dissed me.
...a difference of two clock cycles doesn't appear so appalling when you're running a system with a clock speed of 5000 cycles/sec. Granted that's not today...
No, that's not today, we surpassed 5kHz at least 30 years ago.
Today we run processors in GHz, i.e. 3,000,000,000 cycles/sec.
As an aside, CPU-level functions (like increment and integer) aren't timed in cycles/sec, they're timed in instructions, hence a processor's IPC count. That is what matters, moreso than the clock speed.
AMD has known this for years which is why they try to push their IPC count, as opposed to pushing clock speed.
More cycles, same IPC=higher speed, much higher heat;
same cycles, more IPC=higher speed, marginally higher heat.
But don't get me wrong I like SGI, I like the MIPS It was great for a lot of stuff, First class webservers come to mind first. On the SUN side, they scream with regard to IO. The MHZ issues are somewhat of a smoke and mirrors issue. UltraSparc will crunch data a heck of a lot faster then an 32bit box which is what makes them the thing for ORACLE or simply huge data stores. When was the last time you saw a x86 board that could handle 20 CPUs?
Sorry about the writing. Robot fingers, you know? Cliff Steele in DOOM PATROL #23
Does anyone remember when Sun was the little guy, going up against Dec, Apollo, HP and the other workstation vendors? What was the difference?
MARKETING.
Back then, Sun was CLOBBERING those guys with a better message and machines that were just as fast or FASTER for less money, but the main thing is that they were hammering their message constantly and consistantly every where you looked. They were the upstart trying to get people to think differently about technology and it worked.
The problem with Sun is that they BELIEVED everything they said about themselves to the point of anointing themselves as technical genuises, instead of understanding that they won because they had superior marketing than the other guy. They SAID things that got them attention, regardless of the fact that they built things too.
Now, Sun just builds things and everything they say is directed at the technical people, instead of the BUYERS of technology that decide what gets bought and what doesnt. Geeks may recommend, but they dont to the buying; which is based on more than just who has the better technology. Much of the time it comes down to how much a business can afford to buy versus internal costs and other considerations like THE BOTTOM LINE.
It doesnt matter that every programmer in the world knows what Java is, what it does, and why it is a great advance in both platforms and languages. How many people think they can go into a meeting with a CIO and talk about how Java's garbage collection is going to help his company become more profitable? Maybe it can through cost reductions in programming projects, but when is the last time you saw SUN SELL THAT?
I have never seen them make the BUSINESS CASE for their technology beyond it's geek-appeal, and that is why they are dying. At Sun for the last 10 years, the message has always come from it's Engineers, which is totally and completely wrong.
At Microsoft, everyone understands that no matter what we come up with, if Marketing cant sell it, it gets shit-canned (except MS Bob, thanks Melinda) If Microsoft marketing can make a company believe that radioactive rocks in their lobby can improve their bottom line, guess what is going to be developed? MS Radioactive Rocks, thats what.
Sun has had it ass-backwards for a decade, and time has run out on them. Great technology that cant be sold has killed every one of Microsoft's competitors in the same way. How long a head-start did NOTES have over Exchange? Netware over Active Directory? It's the same old story. They pitched to the techie, while Microsoft convinced corporate management that you dont base your business on field level replication, as much as on messaging and calendaring.
This is the same reason that Linux has NO CHANCE to beat Microsoft in the near future. Linux may be wonderful for those of us who want to invest the time in learning to use it effectively, but most people dont want to face the reality that this group of people will NEVER be the majority.
Most people just dont give a shit how it works, they only want write their documents and to get and send their fuckin email with a minimum of issues. That is 90% of corporate computing, and Microsoft owns it.
Sun and others have decided that they can just change what is important in the corporate boardroom, and they cant. The only company that ever had a real chance against Microsoft was/is Apple.
Why?
FUCKING MARKETING, thats why.
This area is the last large market segment IBM mainframes has, where they are the only player, so this is a serious threat to the IBM mainframes and therefore to all the services&support contracts, and peripheral systems that comes with IBM mainframe ownership.
The recent 100+ CPU servers from SUN and compatible Fujitsu machines as well as their mid-range machines with "hot-swap everything", and everything possible done to make software running on them 24x7x365 capable even while the hardware and OS is being upgraded, is another area where SUN is fast becoming a serious threat to the marketshare and market dominance of AS/400 and mainframes from IBM.
For these reasons alone it would be a very smart move if IBM were to acquire SUN, because it will remove a very serious competitor for from the marketplace.
The new owner (if any) may acquire the company, but will they get Bill Joy as part of the deal?
Chip H.
As a Sun employee, I really hope this is just a rumor and stays a rumor. When Sun merged with AOL via Netscape and then formed iPlanet and then seperated from AOL and Netscape, things were a mess for a couple years and then some. People who should have stayed were let go. Horrible changes in policies (for employees at least) were made... it was just suck all around.
I love working for Sun and I love doing what I do and always imagined I'd be working there for another decade or two at least. If a new ownership means getting laid off no thanks!
Sun's market cap. is 11.9 billion dollars.
Apple's market cap. is 5.82 billion dollars, about half that of Sun.
There are certainly advantages to technology deals between the companies -- both use openfirmware , after all. Rumor even had it that inside apple native versions of MacOS "classic" were built and running on SPARC hardware, and Sun released a version of Solaris 2.5.1 that ran on PowerPC. But I see a merger as being unlikely. Although Quartz on Solaris would be fun!
If Sun is going to be bought, the only deal that makes sense to me would be a purchase by Fujitsu. I can't see Sony jumping into the server business, nor Dell going mano-a-mano with Microsoft.
Sincerely yours,
Jeffrey Boulier
...that one of the NT guys who's suddenly gained an intrest in our Sun systems suggested this very thing a few months ago!!
Then again, it is the SEATTLE Times....
Ok, apple invests in software too, but their hardware is also designed mostly by IBM/Motorola, nvidia and gee... Intel (PCI/AGP). Dell sells hardware THEY designed, using common parts, but THEY designed it. Apple does the same.
However, WHY would anyone buy sun? They only company I can think of is... Oracle. IBM can't even buy it without severy interference by the anti-trust people in the government, Dell will not buy it since it will not add anything to their portfolio: their high end systems are designed around Microsoft's software, not unix.
However, Oracle's database is mostly run on Sun hardware, or better: when you buy a Sun E* server and decide to run a database on it, 10 to 1 it will be Oracle. It does make sence in a way for Oracle to buy Sun so they can sell the complete package to high end customers: hardware AND software.
Never underestimate the relief of true separation of Religion and State.
First they almost killed TCL project of John Osterhout. Then they decided to go with Java (I wonder what was their business model?), then they bought serv-side part of Netscape (anyone use it? or yeah! Sun itself!).
I think HP (a successful collector of dying hardware platforms) will buy their hardware business (HP knows what to do with zomby-platforms), while IBM (as the most successful Java developer) will pick their Java (IBM knows what to do with).
The time of proprietary Unix systems is over long years ago. The market is just adjusting to it by merging till nothing is left to merge.
Less is more !
What I have not been able to understand is, why the outsourcing situation does not give me a clear opportunity to move to India.
-fb Everything not expressly forbidden is now mandatory.
I agree with you that the parent should probably read some more books, but Sun also should be looking more to their worst enemy Microsoft how easy system administration CAN be done. I'm not saying Win2k is easy to admin when you set up a large forest with AD, but applying patches etc, it's darn easy. Sun (and other companies as well) should learn that even professionals do not LIKE it when their job is HARD because the 'tools' they have to work with are very low level and effectually do not help the professional a lot. Tools should help the user, not work against him/her.
Never underestimate the relief of true separation of Religion and State.
good point..
-- NeTMoNGeR
You only show your arrogance when you say that Java sucks ass. Java is VERY powerful on the server-side.
even if Itanium can scale up to hundreds of processors, there's no OS that runs on it which can properly handle that many.
Windows Server 2003 on a 32-processor NEC machine beat out Solaris 8 on a 128-processor Fujitsu for price/performance. Also, Fujitsu has publicly disclosed a roadmap for rolling out 128-processor machines based on the Itanium architecture by the end of 2005.
Windows Server scales up, and scales up well. I would count on it following suit.
The coolest voice ever.
I don't think I'd say Apple has a proprietary UNIX. Their UNIX is open. It's the fourth (5th? 6th? Hard to keep track anymore) OSS BSD. I don't think any closed unices are going to survive. Apple played it smart by locking down as little as possible -- just the part that makes the users drool. Developers are happy because it's all open and available and such, and users are happy because it's a beautiful system where you never have to open Terminal.app.
I think Apple has some incredibly smart people and they definitely played the OS X thing right. I don't think it would have worked if their Unix was propietary.
So yeah, other than that niggle (and it is just a niggle), I think your post is right on.
"Whatever can go wrong, will." --Finagle's Law
That's a big YEP. Eventually once all IT jobs are outsourced to those c*cks*ckers in India, we'll all be relics of a byegone era.
I think you're directing your anger at the wrong people. The people in India are just doing their jobs. The c*cks*ckers are the ones in America hiring them.
What I find funny is that outside workers being hired in the US is being clamped down on, while companies are still free to send work offshore. Huh?
.......and wheres Apple in all of this?
You got me on the cycles/sec, I really didn't mean 5,000 cycles/sec.
Beyond that though, I disagree. Yes, when you are comparing processors, IPC is quite important, as an AMD that performs 9 instructions/cycle vs. an Intel that performs 6 (cited here) would be optimal. However, there certainly are instructions in the set which take a larger amount of time to execute than say increment. Certain instructions by mathematical definition need more time to execute than loading a register, IPC is an aggregate average (see here).
Furthermore, those figures are only to measure the performance of processors, instructions, etc. (as you said, CPU-level functions), and my intent was not to compare processors or instructions, rather to give an arbitrary, static time to a canonical language feature in two different languages. In that case, I needn't back up my data with hard, empirical evidence. My point was simply that the difference between the execution times of a simple operation in two languages is small and static, when the machine they operate on becomes faster and faster, the difference between the speed of the two is minimal at best.
--- What
In other news, doughnut mogul Krispy Kreme (NYSE: KKD) has placed a bid to purchase high-end computing giant Sun Microsystems (NASDAQ: SUNW) in order to secure a first-in-line place for the much anticipated ``blade'' form-factor servers with UltraSPARC IV CPUs to control production, shipping, and doughnut-glazing processes. Officials at Sun Microsystems could not be reached for comment, but a guy hauling garbage out of the rear door of building 7 of the San Diego campus was quoted as saying ``I really like them doughnuts. If they buy us, maybe we'll all get doughnuts for free at lunch. That'd rule.''.
Look out Dell and IBM, when you've got a million doughnut and coffee addicts for your clients, switching them to Unix and Java can't be far behind.
Pining for the days when The Glorious MEEPT!!! graced SlapDash with his wisdom.
That is because IBM is run by adults, who dont need to stroke their ego with reactionary applause with their latest Microsoft insult. McNeely's ego has destroyed his company.
When Microsoft made a better VM for Java than Sun, he should have embraced and extended it. (where did I get that phrase?) Instead, he cried NO-FAIR!!! and went to court. He saw it as an insult, when he could have seen it as a confirmation of Java's potential. He kept his primary technology off of the majority of corporate desktops because he could'nt take it that Microsoft might get some of the credit.
He should have been fired, and the fact that he wasnt, means that Sun's board is a bunch of idiots.
IBM is the ultimate in pragmatic operations.
They will make sure that Java gets back into Windows whatever it takes. You will see REAL Java as a standard language in Visual Studio.NET in exchange for C# in IBM IDE's. They will COMPETE instead of bitch and moan like Sun.
All this "Microsoft is the devil" bullshit will end. IBM understands that business is fucking BUSINESS. It aint pretty. This might even be the opening for Microsoft to start building a few things for Linux similar to what they do for Apple, only in a way that does not cut into their own sales.
Sun was run by children who see things only as "US vs. THEM". When IBM takes over, they will bring maturity to Sun's offerings, by not being such babies about it. It will be good for IBM, good for Microsoft, and most importantly it will be good for Corporations.
Sun is aquired by IBM. IBM releases all of its code under the GPL. Java is now open. Solaris's disk management is now open. Linux takes in the code. The community sets up base in java's source code.
.NET in development speed and features. Crossplatform becomes speedier and more portable. MS no longer needed. Grandma's checking account is now safe from .NET's wannabe all seeing eye.
IBM and the community improve java ala OpenOffice. Java cuts through
I like the Solaris OS, just like most people here. It's stable- not as stable as BSD(netcraft-wise), and a little more stable than linux. But Scott and friends lost their business focus 2 years ago.
To quote a bruiser in pulp fiction- "The world is full of unrealistic mothafucka's. That pain you're feelin now- that's Pride fuckin with ya. You suck that shit up." Scott hasn't gotten a grasp on the reality of his situation. And so now somebody else has to.
Semper vi
The point is that for all of the bullshit that comes out of Apple, they know what makes people buy things.
Apple just never had the dollars, nor the large enough audience for their products. They took essentially good home and school products and tried to sell them in the boardroom for the same reasons.
They are great marketers, but sometimes they sell to the wrong person, instead of developing what the client actually wants. Steve Jobs can sell ice to an Eskimo, but if that Eskimo wants BLUE Ice, then Jobs is fucked.
Apple = Great marketing, bad products for corporations, could be fixed with lots of dollars.
Sun = Shitty marketing, great products for corporations, cant be fixed with any amount of money.
Get it?
Apple (well NeXT) solved that problem already. OpenStep (cant both to capitalise it correctly) ran on both 68k and x86 machines, and OpenStep software could be compiled to be 'fat', ie including both 68k and x86 machine code so it could be installed and executed on either architecture. Wasnt very popular, but disk space wasnt as cheap in those days as it is now.
Apple must still have that code lying around somewhere.
I use Friend/Foe + mod-point modifiers as a karma/reputation system.
First Apple buys Sun, and then we get the whole "switching from BSD to SysV" thing happening ALL OVER AGAIN.
Thanks, but no thanks. That's one piece of "history" that can remain in the past, thanks.
Sun Micro has a grand vision and real scientists in their team. A company like this cannot be bought.
You have a point. Marketing is the most important thing with management. In regards to linux though, it has more marketing now than it ever had- ever.
Look on Gartner Group, CNET, CNN, FOXNEWS, Wired, etc on any particular day. You will see Linux staring back at you. Linux has been proven. It's not another bitch like Sun. It's a fucking vampire, and Microsoft is the main course. This is the new reality. Linux's resilience keeps putting it back in the news. While other vendor's float by face down, linux will remain- and manager's understand this now.
Lets agree to disagree on this one. I'm lucky enough to be in the position to make this happen where I am. And I am making it happen. Although I'm not the only one.
Parent is an excellent post.
I'm wondering if Sun is trying to change who they market to. They are pushing N1 as a means to reallocate resources based on business requirements (see also post further down), and Project Orion to provide regular delivery and simplify software integration as well as new pricing models.
(my take, anyways)
I think this might give their marketers some ammunition. They still need to deliver on these, though...
So does Anonymous Coward have good karma?
The tech industry consolidation cycle is continuing. Intel won the 32 bit war with marketing dollars and superior manufacturing. The 64 bit cycle is becalmed in a fog bank of ennui, although of the major players, Sun which is the most dependent on its own technology has the fewest options.
Every industry consolidates as it matures, and computers are no exception. At this point, when the real competition in Sun's commercial space is whiteboxes running Linux that cannot be under priced and IBM handholding that cannot be out serviced, the time for Sun to know when to fold 'em has come.
Sun's value to a buyer is not technology. Nobody needs it. Intel, AMD and IBM will keep the world supplied with 64 bit chips, and software is even more abundant. Its cash is worth just that, not more. And its personnel, well no hard feelings please, but there are a lot of unemployed engineers out there these days.
Sun's value is its installed base of hardware and its sales force and their customer relations. A rational buyer will sell off or shut down Sun's technology, and use the sales force to migrate the customers over to the buyers products. Without the R&D and the corporate overhead, the deal will be a win even if only a minimal number of customers are retained
I think that HP is the Company most likely to do this. IBM has too many consent decrees and the PW consulting acquisition pointed the way they wanted to go, which is not hardware oriented. Dell, which is a marketing channel, not a computer company would have to keep too many bits of Sun around to make the deal really pay off. The same goes for Apple.
Giving Fortune 1000 companies all the cheap, slavish, technologists from Asia their inner-Raj could desire is like giving high school kids methamphetamine at a pep rally.
Seastead this.
Because Sun has $12Bn in market capitalization and $5.5Bn in cash on hand, I think the question isn't who's going to buy Sun, but rather who should Sun buy?
I have maintained for some time that Sun should purchase RedHat (current market cap. approx. $1Bn if my sources are correct), go whole hog into promoting Linux, move the advanced features from Solaris into Linux, and turn their hardware into the best darned high-end Linux servers and desktops you ever saw.
First of all, IBM is already trying to do this to Sun with high-end servers. New action is needed to defend that ground.
Second, putting the weight of Sun and the open source devotees behind Linux application development together can help cut into Microsoft's server market share and potentially even make some more desktop inroads.
There's probably no getting Sun out of the hardware business. But unless they harness a mass movement behind the software needed for their systems, they face the prospect of being the Apple of the UNIX server world: well-regarded but largely unused.
Linux does have tons of marketing, but that marketing is supported by us geeks who love the fact that we can still get meaningful stuff done on 486 class computers. That is actually how I got into Linux myself; I had 3 486's that I was going to toss, but now they make up the DMZ of my lab's network. That is an effective marketing message; low cost is ALWAYS a good thing to sell.
However, it can only go so far inside the Corporation, because at the end of the day, it has got to do EVERYTHING the competion does; not just do some things more reliably than Microsoft.
Linux will succeed only when two things happen; it gets a Directory service, and they come up with something AS GOOD as Exchange Server. The Exchange Server/Outlook combination is what keeps Microsoft in play in the BackOffice.
It may suck, but it SUCKS LESS than anything else for messaging, and any reasonable person will tell you that.
You wont get those products under OSS because those would be huge undertakings by people who want to get paid. It may happen, but not soon enough for Microsoft to get a secure share in the server space.
The marketing cant be "We are different, and we are cheaper", it has to be "we are better AT WHAT YOU WANT and we are cheaper".
Sun has to change their whole persona. With McNeely in place, any changes they make will look like desperation.
Only getting rid of him, will convince people that Sun is serious about change.
There are plenty of things wrong with Java. But unless you have significant amounts of experience in other languages, as well as a fair amount of understanding of theory, it might not be obvious.
Suffice to say that in terms of language features, Java is one of the least powerful languages currently in use. Both in terms of expressiveness (can you defend using public static final int instead of enumerations, or having to create a new class just to return multiple results from a method?) and in terms of fundamental features (first class functions are missing, generics are missing etc.).
The design seems to be based more on the personal preferences of the designers than on any sound and consistent design philosophy.
Fujitsu is a world-wide manufacturer of computer systems, including the ownership of Amdahl and the competing products to IBM's mainframes that Amdahl has produced for 30+ years. Fujitsu also produces and markets the the Sparc64-based PrimePower series, which do pretty well from a price-performance point of view viz Sun, and scale to the same levels as Sun's big machines. At this time, however, they have decidedly smaller market share than Sun even in Japan. If Fujitsu sees IBM as the long-term competitor globally, then locking down this front and buying into a readily-compatible customer base might make sense. (Some PR on Fujitsu Sparc roadmap here: http://www.ftsi.fujitsu.com/doc/press-releases/200 30312-001.pdf)
The less said about Mike "The Diesel" Spindler, the better...
No one who suffered through those years (they included a significant period when John Sculley seemed more interested in learning to be a visionary than he was in running the company) would say that Apple had "great marketing." Apple has survived because of loyal customers, a commitment by its employees to doing great work on behalf of those customers, and its willingness to continue to innovate in the personal computing market.
But "great marketing?!" For the longest time, the best Apple commercial out there was a Saturday Night Live parody.
Whenever I say anything good about Apple, I am only talking about the Steve Jobs years. Any of that time in between was a lesson in how not to do business.
Your company has nothing going for it. NOTHING.
.NET mentioned.
Nothing but a bunch of sweet military sales and service contracts. At $massive_workplace_machineroom, Suns outnumber Intel machines 2:1. A project recenty added a Sunfire 15000. I wouldn't say that Sun has nothing going for it.
You've got no marketing; In the media, your N-1 gets mentioned once for every 100 times that
That's the general media for the general public who play with Intels, not technical journals that are read by professional CTOs who run shops with > 10,000 users. Besides that, if the media mentions that dog poop tastes great 100 times for every time pizza is mentioned, does that make dog poop taste better than pizza?
Oh, by the way, make sure you do that resume in WORD, so that people can actually READ it.
<clue>
Not if you want a job in a Unix Shop.
</clue>
;)
/*drunk.. fix later*/
Well here's the basic problem with sun micro:
They first started as a mainframe company, longterm very solid slow growth investments. They then thought they'd go into the software business and made Java wich say what you want about it is in many many ways WAY ahead of it's time interms of the technical side but they droped the ball as it wer, the didn't push for a Java PCI on board chip to get at the longterm investors. the computer market isn't and wasn't ready for something like Java/Pine. For two reasons the game people and OS people don't make much on something that's elogant, it's true if Windows worked right from start to finish bill would be worth 75% of what he is.
Sun then thought they might become a internet firm wich is closer to what they started with this worked great untill speculators hypersaturated that market
So their two major markets are worth less, hence so are they.
In all reality I think if they did merge with Apple it would iether work very very well or very very badly, and not kind of mediocre, they could continue to due the fun experimental R and D. They'd have apples marketing machine, apple could develop a PCI Java Chip, and no it won't change the world but it would how ever offer a much much more accessable reason for chip makers, and software people to want to look at the viablility of how attractive going Java, for real all the way is.
It'd give Sun a place to sell solaris and honestly get them to get their act to gether.
C++ can be easier though. It's main advantage it it's multiparadigm. (That's also what makes it difficult though, C++:Java::Perl:(some really structured scripting language)) You don't need to mess with nasty OO stuff if you don't want OO, and your program doesn't take a hit for OO stuff that it doesn't need
I do not understand the fuzz, especially the reasoning of the so-called analyst, who said "Sun is cheap enough and when a company gets down to that level acquisitions do get increasingly likely". Maybe that guy should check the price of Sun shares from October 2002. They were at USD 2.34 that time, that's about 30% lower than today. Why didn't anybody start buying them then?
Signature deleted by lameness filter.
Java's standard libraries are definitely more developed (=larger, more comprehensive) than those of .Net, but C# is a better language. C# has a whole laundry list of nice features that Java either lacks or has been scrambling to add since the debut of C#. I work in both languages, with much more experience in Java, and I definitely enjoy C# (the language) more than Java.
.Net bytecode (IL) reputedly has designed-in support for generics, tail calls, and other goodies that Java lacks. I understand [rumor] that for years Guy Steele has been pushing for changes in Java bytecode that would allow for good Lisp/Scheme support. Nothing has changed with Java, but .Net incorporated his suggestions.[/rumor] It wouldn't surprise me if this rumor is true because I've seen how interested MS was in gathering requests from disgruntled Java developers in order to make C#/.Net more attractive.
But the fact that I can use Java on all platforms of interest to me, particularly Linux servers and client devices, is why I keep using Java, so I'm working with the JCP to try to add the C# improvements to Java. If I could use C# everywhere I use Java, though, I would go 100% C# without a second thought.
Also, the
I sure hope Mono succeeds.
"Those who have never entered upon scientific pursuits know not a tithe of the poetry by which they are surrounded."
"Java sucks ass", Insightful gush /. moderators
/. moderation we've all come to love. And when I say expert and love I am of course using those words in the negative sense.
"Java is very powerful on the server-side" Flaimbait
Another shining example of the expert
Java is a huge server-side force because it is so powerful. Many very high end sites run on Java. JBoss is constantly in the top 10 downloads from sourceforge, and that's not likely because it "sucks ass."
when you say the stock jumped 12% and keep it out of context it does sound impresive... however it only gained $0.41 when a stock is sitting at 3-5.00 mark, any movement is large...
Thanks to file sharing, I purchase more CDs
Thanks to the RIAA, I buy them used...
You have to consider who would be the best steward of the OpenOffice project. One phone call from the Great Satan of Redmond and the project would immediately lose all of its resources and promotion, if it were owned by Dell or HP. That leaves IBM, who although already has their own office suite, is at least known to be heavily invested in open source.
IBM is also the obvious choice because of its commitment to Java.
Tired of FB/Google censorship? Visit UNCENSORED!
I think your comment was subgenuous!
In case you havent noticed, it doesnt matter if dog poop tastes better than pizza if Sun were selling it. If Microsoft was selling MS dogpoop up agaist Sun Pizza, odds are that Microsoft would win anyway, because Sun cant do Marketing. It's just a fact, not a troll.
I'd say it depends on your definition of easier.
If your definition of easier means novices can do it more easily, then yes, Java is easier.
If your definition of easier means experts can do things more easily, then no, C++ is easier.
Personally, I'll take door number two, thanks.
-1 Uncomfortable Truth
Yeah, explain what you mean when you say C++ is more compact and easier to "write" than Java? As a long time user of both languages I can only imagine that you haven't done anything significant in C++.
If Dell, HP or some other vendor bought Sun, would Sun's funding for OpenOffice.org and GNOME dry up, considering these other vendors are primarily Microsoft shops and the funding could generate a conflict of interest?
...when Scott appeared as "Java Man" on the front cover of Forbes. Reading the article, Scott's business plan was not about making money, but about attacking Microsoft. I didn't see how such a strategy would lead to profits.
The second component is the high-end server business. The servers are reasonably designed and could give any company like Dell, which does not have a significant server presence, an instant entry into the market. Sun's high-end servers do not have good performance, but that that has nothing to do with its server design. The problem has been the UltraSPARC III, IV, and eventually V.
The 0.5th component is the perpetual license to the SCO Unix source code. The only company which would benefit from possession of this license is IBM.
Sun has other components of its business, but they are essentially worthless. First and foremost is the UltraSPARC development team. It basically destroyed the UltraSPARC's future by designing a processor that ranks among the worst in performance for the last 5 years. Further, the disk storage business is going nowhere.
What is likely to happen is the following. Dell and IBM make a joint bid for Sun. Once they own the company, they will spin off the worthless parts of the business, or, like HP, they will simply fire the nonessential people.
Then, Dell and IBM will partition Sun's business units. Dell will acquire Sun's server business and will make some minor modifications to the processor boards in the highend servers in order to replace the UltraSPARC chips with Itanium 2's. IBM will acquire the Sun's Java business and will immediately place Java under the control of a standard's committee. Further, IBM will acquire Sun's perpetual license to the SCO Unix source code.
None of these useful components of Sun's business has any value to Apple. So, Apple would not be a buyer.
Sun has been the force behind so many protocols that we take for granted, but they simply don't know how to put together products. Solaris is a great OS, but look at the package. HP-UX just like Solaris uses CDE for its graphical interface, but has an incredible administration tool called SAM. Solaris' Admin tool and others are simply nothing more than crap that could have been coded by any high-school kid. Still, they have had many versions to improve them, but they haven't. With regards to their workstations, again great technology, but have you tried installing an ethernet card in Entreprise 350. Another overvalued piss of crap. Java is simply awesome, but have you used NetBeans? Dude, the idea of an IDE is to make the user's life easier, not harder. Swing is a great example. A graphical API from a company that simply doesn't know how to provide a decent user interface in their products. Lets be honest, Sun's main customer is the Federal government. Who else can afford truck load of overpriced machines? That's not the way to run a company. You cannot depend on one main customer and expect to survive.
I don't know if the rumor about someone buying Sun is true. but it makes sense. I can only see two people buying Sun: DELL and IBM. If DELL buys Sun they are doom, because unlike IBM or HP their infrastructure is not equiped to handle what Sun has. DELL within 5 years will be gone if it dares to do that. IBM, on the other had, will have a lot easier time incorporating the gains of the acquisition. Unlike DELL, IBM would know how to deal with Java and all of Sun's technologies because they have been using all of them for decades. Plus, they have right away places where to put them. DELL will have to start from scratch moving in so many directions that even years after the acquisition they will still wouldn't know what to do with most of them. Their revenue will simply dry up trying to make real the integration. If anyone can pull it off, it would definitely have to be IBM. IBM would be able to pull it off and make money right away, which is what acquisitions are for after all.
Troll??????????????? How do you figure? Does Microsoft have exhorbatant fees on the licenses?
I can remeber back in the early 90's when Sun was a workstation company. Anyone remeber that? Sun had the best Unix workstations around, they competed head to head with SGI, HP, IBM, Apollo, etc. They just crushed the mini-computer makers back in their hey day as you say. Then one day they switched their poduct focus to "servers". At the time I thought that was a little weird since it seemed to go against their then philosophy of "the network is the computer", and having many networked unix stations together to solve scientific problems. You can still sometimes see this philosophy in technologies like JINI, or the work they did on JXTA. Often times it was like Sun would be saying 2 things. One was that their servers were the best in the world, the other that the future of computing is millions of networked computers of different types(smart devices) communicating to each other with out servers. Of course Sun doesn't make any of these other devices, they only make big expensive servers. Sun made a boat load of money in the late 90's in servers, their future is not at pretty in that market anymore. Maybe it's time to make another switch, they went from workstations, to servers, maybe it's time to refocus on "the network is the computer". On another interesting note, Sun dumped millions into Java, and what do they have to show for it? Has java made Sun any money? I think that history will judge Sun harshly on it's java strategy.
"Number 1) Java sucks complete ass right now."
How absolute BAD are the Moderators when a post like this is "Insightful" (which is totally opinion) and the posts defending Java which state fact are offtopic/troll, etc.
The whole Moderation thing should probably be dropped. This is a total farce.
My team mates and I have been using Linux on our desktops since we arrived at our present position several years ago. However we still can't justify a move from Sun hardware and Solaris. The price point and quality are too good.
If Intel compatible machines would put in OpenFirmware or equivalent, remove keyboard and monitor, and allow power management/console access through a single RJ-45 serial connector at a similar price point we could talk. Sun Fire v100 has this in a 1U for $995 retail.
What I'm trying to point out is that Intel machines have an incredible amount of horsepower but have consistently failed on bringing managability in at a reasonable price.
Further I think Sun is in the course of reinventing itself. They are supporting numerous open source efforts, looking for Solaris to Linux exit strategies, and moving away from proprietary hardware that kept its price point high. Just within the last few generations of hardware look at the changes. Say goodbye to SBUS for PCI, special memory for DDR, standard monitors. Similar to the reinvention Apple has gone through isn't it?
New rule - all posts should have a list of the people that moderated the post and what their input to the moderation was.
Critics should not be nameless.
I don't think anyone here will quibble with the idea that Microsoft puts an emphasis on marketing.
For me the question is, to what extent is this a good thing, especially for the consumer?
It is a tautology that a product can't succeed, no matter how good it is, if it's existence is never revealed. A product that is of no value can, on the other hand be sold to many people with a sufficient amount of marketing.
These two facts may lead some people to think that the budget for any company under any circumstances should favor marketing over product development. I don't see it that way. In fact, while Microsoft has gotten where they are today by out-marketing companies who in many cases had superior products, I think MS is now a victim of it's own success with that strategy. Having dominated the desktop as they have, it's going to be difficult if not impossible to convince the majority of Windows/Office users that they HAVE to have the next versions of those products. In fact most people I talk to who are happy with Windows/Office think that the current versions of both products are TOO complex and would love to have something come along and simplify their PC usage.
Microsoft may be able to sell developers on dot-net, I've talked to developers who seem oh so excited about it, but can't quite explain to me what it is. But the real question is what is the next big money maker for MS on hundreds of millions of desktops? I don't think Microsoft knows, and I don't think they have a line-up of possibilities waiting in the wings either.
When it all comes tumbling down, there will be arguments about whether Microsoft lost its marketing touch, or whether product development simply failed to deliver the goods.
The truth is, that by letting marketing considerations predominate, Microsoft has produced a glittering tower, and in their haste, not provided a solid foundation for it. The sorts of things intentionally left out of that foundation (a robust command line interface, security, hardware portability) cannot be patched up with a new GUI front end (although that won't stop them from trying). What Microsoft needs to do now is what Apple has just finished doing: throw out the underpinnings of the OS and start over. The difference is that Microsoft will almost certainly lose market share in the process, and that market share will be lost in the only areas of the company's product line that are profitable. Had Xbox, MSNBC, Slate or their consulting division efforts been successful it might be a different story.
Regardless of what happens with Sun, Microsoft needs to pull a big rabbit out of their hat right now. I don't even see ears yet.
I can't believe people are actually suggesting Java could dissappear. What does Sun have to do with Java anyway? High speed file host for the JRE and SDK?
Welcome to Java .NET edition (all existing java code is broken, so don't bother).
And our new sister company, Microware will now provide the only working hardware for Microsoft software.
Agreed, as a C++ developer for over 12 years and a Java developer for over 6, I'd have to say that Java is FAR, FAR "easier" to develop with than C++.
Compactness is a bit of a wobbly term to throw around - are the C++ EXEs more compact than a JAR file? No. Are the C++ EXEs more compact once you include the JVM with the JAR file? Yes, but then you have to be fair and consider that a large number of pre-installed libraries your C++ EXEs are dependent on must be included.
Pleasant... no, C++ isn't a pleasant language to develop with. It's syntax is at times absolutely abominable and anything but clear and clean. I think it was DDJ who asked prominent people in the industry to rate themselves based on their knowledge and proficiency of C++. Even Bjarne Stroustrup only put himself at 70%.
I'm not going to get into the "better" argument since it's nonsense. The arguments made by the original poster clearly show that he's never undertaken serious development (on a large scale project) in either language.
I love Java, I love C++, I love a number of other languages... and they all have their place, but Java (and Java-style languages such as C#) are a great improvement on the C-style languages (such as C++) for clarity, readability, development time, and ease of use. It's time to evolve.
As soon as you close your mind to any technology you've pigeon-holed yourself in such a way that you've guaranteed you become obsolete.
of java? I just finished a few hours of searching about the web, and couldn't find a single place to buy even a pound of coffee from Sun. Maybe that's why they can't cut it in the market-- poor product distribution.
Sun totally defines themselves in terms of toppling Microsoft, and that fanatical hatred is so deeply integrated into their corporate culture that they are incapable of even considering doing anything else. If Microsoft disappeared, Sun would dry up and blow away, without any idea of what to do next.
Sun does't develop software or hardware, they develop badly designed weapons for toppling Microsoft, i.e. Java, Solaris and its ilk. Sun views their customers as mercenaries in their holy war, just like the U.S. views the Kurds in their fight against Hussain after the first Gulf war -- they make lots of empty promises about support, hype them up with lies to rebel against their cruel and ruthless leader, then thoughtlessly abandon them to be gassed and slaughtered in the battle field trenches, without any regrets. Sun is much more interested in exploiting the propoganda value of their hated enemy slaughtering their customers, than actually helping those customers.
In this day an age, anyone who's still a Sun customer is just plain stupid. If anyone buys Sun, it will be one of their few remaining loyal customers, because nobody else in the industry has the right combination of suicidal tendancies, unbridled malice, ignorance of history, and sheer gullability. (AOL/Time Warner, Worldcom and Enron come to mind...)
Since any corporation that bought Sun would have to put years of work into de-toxing them from their self-destructive obsession (which would be less possible than reforming David Duke to marry a sharp spoken black lesbian poet), purchasing Sun Microsystems would be an idiotic thing for any company to do.
A patently false rumor about Apple buying Sun ran through the mill a year or so ago, and it was obviously impossible because of the extreme mismatch of corporate culture. Apple has its problems, but they're not that dumb. Sun employees would never get along with Apple employees, because the Sun employees think they're all fucking geniuses who want to dictate the behaviors of people who they consider stupid (Java, java, java! Attacking Microsoft is more important than developing products or supporting customers!), and Apple employees simply won't put up with that shit. Remember what happened with the Sun/Netscape alliance and how much the Netscape engineers resented Sun.
Would any of you Sun/AOL/Netscape alliance refugees like to testify?
-Don
PS: FYI: The Vice President of Marketing for Electronic Commerce of the Sun/AOL/Netscape alliance is a dyed-in-the-wool evangelical Scientologist, who applies Scientology Cult Teachings to (in her own words) "every marketing plan I work on".
Take a look and feel free: http://www.PieMenu.com
This is a badly kept secret, but Sun will be part of IBM before the year is over. Just remember to comeback and tell me I was right when it happens.
But wouldn't it be better if the Open Source community grew up, and got some useful work done instead of bitching about Microsoft all the time?
Zander should have been running that company years ago, and now he's gone.
That link talks about a server not a chip. the E10k is what came out of the cray purchase
the direction of Sun is what's really important. Less and less market share of hardware, poorly performing architecture on the low and mid range, and their refusal to promote and develop Linux on the high end because of fear it will hurt solaris sales.
I see no future for Sun now that Linux has gained the high end enterprise system features in the kernel, and the related high end datacenter administration tools are in the works.
Sun is dying, and Linux is killing it.
Two, regarding .NET; there are only two sales audiences for .NET, the developer and the CIO. Any company that develops for Windows, is either using .NET, or going under. Going from C++ and VB as options to C# and VB.NET is like going from Budweiser to Guinness. Both in productivity and functionality. The .NET Framework classes are outstanding.
The difference is that Microsoft marketed .NET towards creating a groundswell of interest, without having to explain what it is. You can look at the bookshelves at Barnes and Noble to gauge whether or not they have been successful. If you can sell a product without even having to explain it, I would call that pretty damm good marketing.
Java = Great Idea, Sucky Marketing, Barely Decent Implementation.
I dont agree with your "glittering tower" argument, because those of us who write for .NET are estatic with the tools we have from Microsoft. Can ANY Java developer say that about Sun?
What .NET has done, is give Microsoft the time and distraction nessesary to fix Win2K. Windows Server 2003 is MUCH more secure, faster, and with a SUPER IMPROVED IIS. ASP is fixes as well, so no more spaghetti code.
Marketing is not just selling, ya know? Microsoft waved their left hand and you paid attention to it, while their right hand was fixing Win2k. Had .NET not been around to talk about, Microsoft would have had a lot of explaining to do about security. Now, for the most part, security is much improved, and developers got a new language, and a new set of classes to work with.
Pretty smart if you ask me.
Well, that's a few criticisms I've never heard before. I'll give you compact, and if you want it, a performance edge. But other than that, C++ is a much more difficult language to write real programs in. It requires approaching guru level (well beyond expert) to write a portable system in C++. Language feature interactions in C++ can put a bug into the most subtle of indiscretions.
- Operator overloading is ammunition for the inexperienced to do something that looks cool but is actually extraordinarily unwise.
- Const and const correct sound good until you inform the newbie of the mutable keyword and later find that you don't have const methods after all.
- Templates are similarly powerful/risky. Java will get them in 1.5, but the issues around their effective use are legion
- Using threads in C++ is akin to a black art. I used to literally start any discussion of C++ threads by drawing a pentagram on the whiteboard to remind everyone in the room that we were about to descend into the depths of the various C++ threading models.
- Dealing with other people's screwed up multiply inherited class structures was the only time in my life I've had migraine headaches
- How about syntax-dependent semantics for the static keyword?
- And though I tend to prefer the more explicit hpp/cpp interface/implementation separation, method inlining manages to ruin it right away (without any known benefit since the compiler will inline or not without your hint). Also, do I really need to type so much to get the hpp/cpp separation to work?
- You've still got the C-preprocessor. Have you ever seen how much damage someone can do to code readability with the C-preprocessor? It's worth it to move to Java just to avoid dealing with cpp.
- Portability. Java isn't really 100% portable either, so I'm not going to make that claim. But unless you're able to stick to gcc, porting a C++ app to another system is agonizingly difficult unless you're a guru on both systems. Even then, I'd probably get a whole chicken just to make sure.
Now I agree that Java has its warts and there are plenty of aspects about it that should get attention (jdbc, unicode handling, API conventions, faster elimination of deprecated elements, etc.) but it has some huge advantages as well. The first one, one which almost any other language could take and benefit from, is the deliverable packaging strategy. TheAnd here's the scary/exciting part: C# is awfully close to doing all of that. The only thing that C# doesn't really have is the huge library and the mindshare. The C# library will grow. Depending on how it grows (.net-only or portable), it may get the mindshare. Java has one or two chances left to fix itself before momentum starts building behind C# and the future will be a very interesting time. It looks fairly certain that we'll have C# (and Microsoft) to thank for speedier future changes in the Java language.
Regards, Ross
If what you say is true, and I tend to think that it is, then it only underscores the need to do the sort of thing I was describing.
Sun needs to get in front of the Linux parade in a big way to protect its future.
IBM has done far more to popularize Java than Sun. Just look at the Alphaworks web site. In general, IBM has a better grasp of the software world. The two companies actually have a parallel history. The main difference is that IBM has long since worked through their "we own the world" phase. If IBM were to buy Sun, things would get very interesting.
One thing: last time I interviewed at Apple (97) they were planning to replace Objective C with Java as the main system language. The object models are close enough, and there are a lot more Java programmers. Still hasn't happened. Ideology again?
Maybe the smp overhead killed the Solaris performance. There's a point where adding another processor may actually _reduce_ overall performance.
A fair comparison would be to run Solaris 9 on 32 processors also (the same ones, preferably, or equivalent if that's not possible).
--
You seem to be saying is that the language shouldn't provide the programmer with power that can be abused. I strongly disagree; bad programmers will abuse any language feature, and good programmers shouldn't have to do without features because of that.
I'll concede this.
Whereas in Java, you have built-in synchronisation that lets you (dead)lock on any object and encourages you to wake-one when this is very rarely correct.
No worse than final.
It generally can't inline across translation units.
Most of that is necessary to resolve which version of an overloaded function you mean. I think.
This is just nonsense. Writing portable C++ code is easy if you think about the need for portability in advance and abstract out platform-dependencies. GNU C++, Visual C++ and the EDG-based C++ compilers (Intel, Comeau, ARM, etc.) are all very standard compliant now.
IBM gets a NEW perpetual Unix license, a Unix derivitive that has code which can be trasferred into Linux without any possible objection by SCO, and final control of Java, which means they can do what Sun never would; release it as a true public standard. I suppose since they also make PowerPC boxes, another odd and interesting high end hardware platform would not be impossible for them to manage supporting, either...
Now the other choice is HPCOMPAQ, a company/merger known for disaterous mergers and absorptions (think of COMPAQ's grab of DEC and Tandum) that soon devalue the products they aquire to zero, and while once composed of real r&d companies that actually did new and creative things (HP, DEC) are no longer anything but a large purposeless box mover for Intel and Microsoft, and never to be or become anything else evermore.
My hopes are on IBM...
I should not feed the Trolls.
I should not feed the Trolls
I should not feed thr trolls.
I will burn my karma anyway, feeding this troll.
A nice little thing known as a EULA.
Feeding the troll is stupid.
Do you consider giving a third party blanket permission to delete any and all software on your system to be be something other than exhorbatant? Especially if you only find out about it, after the fact.
Wind under Thy Wings
Amber
Suppose you did.
Suppose you did not.
> There is a reason C++ journals have sections
> devoted to obscure sections of the standard and
> how code might not compile the way you would
> expect it to.
There's a reason Java journals have sections devoted to improving performance, decreasing memory usage and how code might not run on the VM the way you would expect it to. Easy? Sure. For writing simple stuff...then you try to make it scale...then you start decompiling the JDK to figure out what is going on...
You seem to be saying is that the language shouldn't provide the programmer with power that can be abused. I strongly disagree; bad programmers will abuse any language feature, and good programmers shouldn't have to do without features because of that.
No, I'm responding to the assertion that C++ is easier to write. It has additional language features that allow for more compact notation and in some cases more useful constructs. All of those features introduce additional complexities into developing a robust system.
I appreciate that there have been plenty of times when I would love to have operator overloading in Java (I was working on a Money class just last week), but I also remember having to unravel some code made nearly inscruitable by someone fresh out of school who thought it would be neat to add some semantics to a few operators because "they're smaller than typing out actual method names". That's a pathological case, but I hope the point is clear. The additional language complexity of C++ does result in more expressive power, but comes at a cost of requiring more expertise.
Writing portable C++ code is easy if you think about the need for portability in advance and abstract out platform-dependencies.
And you're the lucky one if those who originally wrote the code you have to port actually thought about those issues. Total cost of system development is dominated by maintenance costs and Java is much easier to maintain than C++.
I'm just thankful that I don't do maintenance any more and am largely responsible for top level architectural/design issues and development through about the third release cycle. But I still try to make life easier on those who have to come after me and Java makes that much easier to do.
Regards,
Ross
Actually I was thinking more along the lines of make twice as many moderator points available, but require 2 seperate people giving the same moderation before it even shows up. I like the posting of the moderators names next to each moderation too, though.
I doubt the second Insigtfull moderation would have been done if he couldn't have seen the first moderation.
Is there a preference to turn off moderation?
One last request, I think anybody mentionaing moderation and karma should be automatically modded down, including all of these posts. Half the posts I read now are talking about moderation and Karma. Its become an obsession to the point we cannot read the posts! Seriously.
Seriously, no one would want Sun in our current configuration. Too many idiots have spoiled too many projects.
... there are just too many dorks trying to steer things in weird directions instead of paying attention to the market and competition.
... now it is finally out and we're barely keeping up with what everyone else had time to release during our twiddling. Almost purely because of internal politics.
But, if we can finally get things turned around, you better buy fast because we will regain our strength quickly.
The capability is there
I recently finished a project that should have released months ago and would have been on the top of the curve
Obviously I'm speaking my opinions of my employer, not my employers opinions. Also obviously my opinions are why I am posting anonymously.
The question, of course, is not how many CPUs the OS can address, but how much you gain from adding a CPU.
In an Oracle RAC Cluster with Gigabit interconnect, you start to saturate the interconnect bandwidth with over 4 machines, so the "128xDell+Linux" solution Oracle are pushing is bogus. Also bear in mind that 1xGB connection will fill a single CPU in itself, so with a 4-CPU box, you now only have 3 CPUs per box in a 4-box cluster.
It's not just a numbers game, it's a scalability game. Linux isn't scalable beyond 4-8 CPUs, Win2k beyond 4, Win2k3, according to you, can support 32 CPUs (ooh!), and beat Solaris 8 on non-Sun hardware. Thanks for the URL, by the way ;) Who did the survey? How did they set it up? Why did they not use Sun hardware? Oh, just insignificantly, what was the app? Samba?!
I have Win2k Pro on my laptop; I've used it about 5 times (max 3h per session), of which it has crashed within 5 minutes of booting twice. With no non-MS software, on a Dell C640 laptop.
Oh yeah, I'll trust some geek on /. with no URLs saying "32-CPU Win2k3 beats Sol8 with 128 CPUs"!
Author, Shell Scripting : Expert Re
Of course, there was a standby PSU, so the machine could still be brought up, but it highlights the stability of these boxes (and that box was probably 3 years old 4 years ago - lessons like have been learned, and things have been improved since then.)
Just the shock of being rebooted gave the server a kick up the backside. Of course, the PSU was replaced within hours by the support contract.
That was, IIRC, a 4-CPU box. Give me an example of a 4-CPU x86 box which was:
a) In existance 7 years ago
b) Only rebooted once
c) Running a business-critical app for a major UK insurance firm
Bonus points if it runs Windows!
Author, Shell Scripting : Expert Re
Stuff like Java helps to sell servers.
Author, Shell Scripting : Expert Re
Hahaha. This is hilarious news. Imagine if Microsoft bought them out.
Sun's strongest market is in the mid to high end server space. The most important innovations they have made there in the past couple years include: 1) NFS and NIS+ 2) x86 Solaris These are excellent technologies and should be applauded. However, they effectively give Sun's customers an elegant and relatively painless way out. Using Sun's own technology a company which was previously dependent on Solaris and SPARC hardware can switch to commodity hardware and linux first on the desktop, then on the low to mid end server, and finally on the high end by replacing SMP SPARC systems with Linux clusters. Java is not going anywhere, but it also fails to make Sun any real money. And, despite some of the silly bickering between Sun and Microsoft over Java, Java actually makes money for Microsoft. A lot of Java development happens with Windows both on the desktop and the server, and Java's vendor independent persistence technology makes Microsoft SQL Server as attractive a platform as any other (Microsoft caters to this by providing free JDBC drivers while at the same time maintaining vendor lock-in for .NET persistence.)
You get a crappy admin (or two if we include you), Sun reccomends a cautious approach (because your configuration is selfinflicted crap) and Sun is suppossed to make it painles for you?
I am sorry but bullshit. The Sun chaps are far to nice to you, any admin worht his title knows that judicious patching is required to avoid this mess.
The only thing more mindshocking is that people in this forum are modding you as interesting or insightful.
IANAL but write like a drunk one.
I didn't mean that Microsoft owns the server room right now. I was explaining my (conspiracy) theory on how what I think Microsofts strategy is to own the server room _in the future_.
I think that they are _almost_ in a situation where they can dictate what protocols and what services a server must provide, due to their control of the desktop.
So that was the idea in my previous post, the only other entity I know that would have a desktop/server combination powerful enough to challenge Microsoft (considering maturity, money and usability) is the Apple/Sun combination.
Linux with all its backers may get there in the future butit isn't there yet. Not until it has a strong foothold on the desktop.
Just my 2 cents.
The Internet is full. Go Away!!!
Guys: do you have an interest in Sun been bought or do you keep pushing these unsubstantiated rumours out of the goodness of your hearts?
I mean, honestly, you have been pushing this crap about Sun for several months now, every single snippet found by somebody about Sun's demise is dutifully put in the frontpage of this well loved, but some times ailing, site.
What is it with Sun that opens a wound with you?
Inquiring minds would like to know.
IANAL but write like a drunk one.
OF course Microsoft wants to own the server room, and the living room, and every other room, but they don't yet.
But if you need a large memory bandwidth, I think probably still beats out Itanium, and definitely beats x86.
That's where you're wrong. x86 and Itanium both beat Sun on memory bandwidth - and not by just a little - by a LOT. Take a look at supercomputing applications which require tons of memory bandwidth. You don't see supercomputers getting build out of SPARCs. They're Itanium, x86 or Power architectures.
If you need a whole shitload of CPUs in one box, Sparc is also a better architecture - even if Itanium can scale up to hundreds of processors, there's no OS that runs on it which can properly handle that many.
Now that Opteron is out in the wild, x86 can scale just as well as SPARC can (thanks to Hypertransport). All we need now is a system builder to build it, spec it out, make it a standard, and let the Dell, IBM, HP and whomever else wants to (white box servers...) build them up and compete, just like the PC market. Do you think that Microsoft or the Linux people (IBM mostly) are standing still? It's only a matter of time before the OS gets the features needed to scale up. There's simply too much money involved, and several fierce competitors to let Sun have the high end market.
Standardization is moving up the enterprise stack. Systems are getting spec'ed out to use standard CPUs, standard components and standard software. This will kill Sun's vertical business model.
That is, when reliability is taken into the equation. That's why businesses deal with SAR (Servicability, Reliability, Availability).
60*24*365=525600 minutes/year.
90% = 52 minutes downtime per year.
95% = 25 minutes downtime per year.
99.999% = 5 minutes downtime per year.
You have the option of buying cheap kit, and hoping that a Linux/x86 box will run with less than 1h downtime each year, which will get you nearly 90%. Of course, you've got to get your hardware support contract to give you that, and the 24/7 software support to tie in with that.
For 5 nine's, that is, 5 minutes downtime per year, which is a very significant cost for big firms (think Amazon.com, or your local telco or supermarket - they could lose millions in five years). The difference between 52 minutes and 50 minutes if very significant.
Author, Shell Scripting : Expert Re
This article is a nightmare to the storage world. No other unix boxes in my experiences has delivered more reliable data I/O bandwidth. Seriously, X86 processors may fly by the $$$, but any mission critical jobs are better left for unix machines.
IBM hardware is good, but their software is overly complicated. Linux hardware is just no reliable on x86, but their software is ok for the storage world. M$ has great software, but x86 architecture doesn't cut it. The list goes on.
The only company with the real median was sun. The Enterprise systems have excellent bandwidth all around. Good for SAN, backups... you name it. Without sun, veritas, legato and a host of other storage companies will take five steps back.
To find out about Win2k3 performing well, trys ults. asp?resulttype=noncluster
http://www.tpc.org/tpcc/results/tpcc_perf_re
I have heard (although not personally verified) that Windows is now the number 1 on *every* performance test on TPC.
Read reviews of shopping cart software
> The Scientologists don't think they're a cult, either.
On the other hand, the Scientologists could probably design a working OoOE processor.
*rimshot*
'jfb
To spur "enterprise Linux," Big Bang, the distributed two-phase commit.
Don't forget Scott McNealy's great innovation
called Sun Bid. Every Sun customer was expected
to bid on the server. Highest bidder wins.
Oh yeah. That went under after millions in
hardware, software and programming hours
went down the toilet...
Did this article have a picture attached showing a woman holding a coffee mug?
ISVs make money selling into the hugh Sun installed based. Customers buy the solutions from the ISVs and select the platforms that are well represented in that ISV's installed base. For many large ISVs, Sun - SPARC - Solaris is a tier one platform and often quoted by ISVs as their reference platform. The on-going Sun revenues of $12B/year keeps the ISVs in the game.
Smart ISVs can keep their code portable, but there is a non-trivial cost for supporting each additional platform. With IBM and HP less committed to their brands of UNIX, ISVs know that they can still make money with Solaris. HP and IBM still have significant installed bases for AX, HP-UX. ISVs still support those platforms in addition to Solaris. By the way, IBM makes a ton of money selling their s/w (e.g. WebSphere, Tivoli, Lotus and DB/2) on SPARC/Solaris. As does HP (e.g. HP OpenView)
Systems Integrators drive a lot of business in the commercial and government spaces. Most of their revenues come from selling solutions that consist of services and products. SIs like Sun because Sun does not compete with them.
Resellers also drive a lot of business. Sun has a healthy channel program and it's partners drive business by selling solutions.
Companies who buy large systems (e.g. >$100k initial purchase price), look for many things:
Multi-billion dollar year firms have tried to implement large complex business systems using tier one s/w, SIs and platform vendors and failed - and in so doing have caused those business to go bust. Many firms are engaged in implementing such mission critical systems and IT professionals know that buying the right ingredients along does not guarantee success.
Sun is clearly a leader in enterprise computing. When large deals go out to bid, the list of tier one plays has shrunk from the old BUNCH days to IBM, HP, and Sun. (Fujitsu and Unisys are tier 2)
Sun is on of the few remaining major computer systems companies. It has a rich set of products, service and partners. Any firm that continues to spend billions of dollars a year in R&D and creates innovations like Java should not be dismissed lighly.
John J. McLaughlin, Editor-in-Chief/CTO, System News Inc. Publishers of "System News for Sun Users"
i thought we were moving *away* (if very slowly) from 8086 technology!
.
i mean, 386 was a big step, 586 a bigger step, itanium an even bigger step... x86-64 - a smaller step, but still away. .
(unless time has reversed itself)
--TRR
You are comparing a box a yet to be released box with 1.5ghz Itanium2 (EA 10/22/2003) to a box with 563Mhz CPU released in 2001.
How is that indicative of OS performance?
Also 2003 is brandnew solaris 8 was released in 2000. Solaris 9 has quite a few performance improvments than 8 so I think the comparison is unfair on all counts, so please check your facts before you posts.
I attribute this mostly bad teaching examples, and lack of experience. One of the most important things to realize about programming in C++ is that you must learn how the features work. I personally don't use op overloading unless there is an "obvious and expected behaviour" to be implemented.
Const and const correct sound good until you inform the newbie of the mutable keyword and later find that you don't have const methods after all.
By using 'const' you have made a contract regarding the visible state of an object. 'mutable' is a statement that you are allowed to violate the letter, but not the spirit of that contract wrt a single member. No black magic there.
Templates are similarly powerful/risky. Java will get them in 1.5, but the issues around their effective use are legion.
There is definitely some tricky stuff in the standard re templates, and every vendor implements those parts differently. I've put off using templates in portable code for at least another 6 months.
Using threads in C++ is akin to a black art. I used to literally start any discussion of C++ threads by drawing a pentagram on the whiteboard to remind everyone in the room that we were about to descend into the depths of the various C++ threading models.
Threads are tough in any lanuage, but Java's portability is a nice change.
Dealing with other people's screwed up multiply inherited class structures was the only time in my life I've had migraine headaches
Those people are wrong. MI is very rarely justified.
How about syntax-dependent semantics for the static keyword?
I chalk this up to a strong desire to avoid adding keywords, every new keyword causes problems with old code. Also, the different meanings of 'static' are not likely to cause confusion in real code.
And though I tend to prefer the more explicit hpp/cpp interface/implementation separation, method inlining manages to ruin it right away (without any known benefit since the compiler will inline or not without your hint). Also, do I really need to type so much to get the hpp/cpp separation to work?
and it can cause problems in shared libs (in addition to those brought on if you use templates). I pretty much agree here, leave 'inline' to the compiler.
You've still got the C-preprocessor. Have you ever seen how much damage someone can do to code readability with the C-preprocessor? It's worth it to move to Java just to avoid dealing with cpp.
I wouldn't go that far, but I do hate dealing with cpp damgage. cpp macros should be prefixed and used sparingly (to do things the compiler cannot).
Portability. Java isn't really 100% portable either, so I'm not going to make that claim. But unless you're able to stick to gcc, porting a C++ app to another system is agonizingly difficult unless you're a guru on both systems. Even then, I'd probably get a whole chicken just to make sure
But you can often be "portable enough".
C++ has plenty of problems, but Java will not replace it for many applications that have limited portability requirements and need very high performance. The fact that C++ is not especially easy to use will not change that.
My main comlpaint with Java is a specific attitude: "If one level of abstraction is good, then 10 must be even better". This can cause more of a portability headache than overuse of #ifdef. A careful C++ programmer will often move portability problems to a single module, and require a new version of that module for new systems. This acknowledges and deals with the problem, rather than adding additional translation layers that can kill performance or lead to programmers using non-standard hacks to avoid those performance problems (or vendors adding non-standard extensions for the same reason).
[Set Cain on fire and steal his lute.]
The only thing Sun has is Java and IBM's kicking their butt there.
Get a clue, or dont bother with this.
SGI has a 128 CPU machine on Itanium2
-- Leeeter than leet
Sun emphacises instruction bandwidth over multiple threads, rather than single-thread performance. Makes for bad benchmarks, but good overall throughput (same idea as IBM mainframes, which have slow CPUs, but giant I/O bandwidth that dwarfs any bus-based sytem).
All the things I am reading these days are very reminiscent of the days in the '90s when Apple was in the dumps. Back then everyone was saying Apple's days were numbered, the company would be bought, the company would go bankrupt, etc., etc. Like Apple, SUN has great technology and a golden brand name. Sun's real problem is they have lost their way and their business model is floundering. An example: One of the SUN administrators at my company was complaining of SUN's current bewildering product lines. There are too many models and many of them overlap. Just like Apple. One of the first signs of the turn around at Apple was when they began streamlining their products into clear low- middle- and high-end machines. I'm a big fan of Java and I don't want to see SUN go under. Maybe they can borrow Steve Jobs for a couple of years.
...Disney! :)
Liberty uber alles.
Ross,
I agree C++ isn't a toy. But, then, isn't it sad that we have to cripple everything to accomodate so called "professionals" who are unable master the tools.
Imagine a world where your Baker didn't now how to use the oven. Or the carpenter that built your house didn't know how to measure. Or the auto guy that doesn't know how to fix your brakes. Making the "standard" cold flour paste, straw huts, and bicycles just seems wrong to me. But, that's what we accept in IT.
C/C++ are fine tools. To cripple an entire industry to accomodate the least common demoninator because "Operator overloading is ammunition for the inexperienced" is just such a very sad commentary on the state of management today.
It seems Sun is going the other way with it's Orion server licensing system; they want to push Solaris on x86 more. Well, if a few truly huge corporations can buy into this, might even work. But it seems more likely they're going to miss out on the Linux bandwagon.