Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Mary Cluney who wrote (121106)	12/7/2000 1:53:08 PM
From: The Duke of URL©	Respond to of 186894

Nice post.

To: Mary Cluney who wrote (121106)	12/7/2000 2:07:06 PM
From: Dan3	Respond to of 186894

Re: there will always be creative CIO's finding a competitive edge by being first and getting there much faster with all the technology at his disposal - even when that technology is not yet available. This is what the future of global competition is all about. The cpu, no matter where it resides, on the client or on the server side, will never be powerful enough. Very well said, and you included the business case, too. Regards, Dan

To: Mary Cluney who wrote (121106)	12/7/2000 3:12:48 PM
From: WTSherman	Read Replies (4) \| Respond to of 186894

<First of all, I doubt there was ever a 30-36 month replacement cycle. Where do you find this metric? Anecdotally, I would guess the replacement cycle is more like 60 months.< 60 months? Not even close. That would mean that on average the the standard PC of 1993(a 486DX @100 Mhz, with 200MB HHD and 16MB of RAM) wasn't replaced until 1998???? By 1998 you couldn't even give these machines away to schools or charities by 1998. The 36 month cycle is from a Gartner Fortune 500 survey of IT directors that was done in 1998. The estimate I used for current replacement cycles of 42-48 months is my own perception based upon working with a large number of different company's. < The vast majority of the replacements are those character based applications running on dumb terminals using mainframe computers running mostly cobol programs that are still in place because the graphically rich and database intensive replacement programs required processing capabilities that are still not cost effective enough to replace. < The only people in corporations that have been stuck with terminals are a limited number of clerical workers. EVERYBODY else has been using PC's for at least 4-5 years. If they worked with a mainframe app they used terminal emulation. But, in the past three or four years a huge number of mainframe apps have been replaced by C/S apps. Y2K paranoia accelerated this transition dramatically. <The overwhelming majority of Corporate purchases are for applications that have not yet been developed. Go to the vast majority of Corporate IT departments and you will find backlogs of 4 or 5 years worth of applications that still need to be developed. That is not to mention millions of man years backlog of improvements to existing applications.< I'm not sure what you are saying here at all. There is plenty of app development or improvement on the IT plate, for sure. But, new apps have not been the driving force behind desktop replacements, ever more bloated and complex desktop software has been. If you don't see that replacement cycles are lengthening I think you're not seeing the forest for the trees.

To: Mary Cluney who wrote (121106)	12/9/2000 9:51:23 AM
From: Joseph Pareti	Respond to of 186894

cool. Mary has got the cool post of the day. It's not again the next bear that cuts and pastes some stuff from 1929 or some clown that ascertains M$FT PE=6. It's an Intel investor ( I would assume LT, please correct if i am wrong). We might have different views on details (mary and I surely differ <g> ) but the KEY POINT IS INTEL INTEL INTEL If I were short Intel I would start worrying, think about... last year or so the "cool posts" were echoing Merry-go-round-meaker and godzilla.com, then there was a turn and the cool posts were ... oh well I have said it too many times

To: Mary Cluney who wrote (121106)	12/11/2000 12:06:50 PM
From: Scott Bergquist	Read Replies (1) \| Respond to of 186894

Having gone from all dumb terminal work in the 80's to dumb terminal emulator on the PC's in the 90's, I have a fair view of big corporate plays in the California arena. Here are two factors unmentioned: 1)e-mail, and the corporate topology of e-mail, is often what drives hardware replacement. PROFS and other character-based e-mail is always a target of GUI-oriented managers. As performance with e-mail suffers, or becomes less compatible with an older platform, the surest fix is to upgrade to a new machine, rather than find a software solution. This is a result of the proliferation of programs, especially Java applets, on any typical machine. Rather than unravel each user's concurrent pile of applications, god knows how modified, it is easier to buy new replacement machines, as new e-mail software is brought online. 2) More and more, large corporate architectures are server-based, and adding more servers, rather than replacing existing servers with more powerful ones (though both episodes take place) is the more common history. Even for the most basic reports and calculations, users far more often enlist expert help (ie, in the IT dept), then accept a "paper printout" at their printer, or by "snail mail". If you are piling up all the ATM transactions at a bank, or polling ATMs to see if they're active, this is where you put your CPU power. But getting faster at that, than the speed of polling and compiling in 1997, adds no advantage. Only aerodynamics, and hydrodynamics, etc are helped by more CPU power. E-mail users are SCREAMING--NO MORE UPGRADES!!!! extra note) The only people needing a "graph" in reporting, are the annual report people. Daily knowledge, reporting, monthly reporting, never relies on graphs: the =real numbers= are alway required, and the "graph picture" is just a "pretty thing" added on, and ultimately discarded (I have found). I find the idea of the average corp user even doing actual computing (other than pixel-math for a picture) more than once a day avg., an exaggeration.