Comdex: Harbinger of the Net to Come

greenspun.com : LUSENET : Poole's Roost II : One Thread

When the Justice Department told Microsoft they would have to split into two companies it was commonly believed here in Microsoft land that it wouldn't matter because operating systems were not going to to be worth much in the future because we would be going to web based systems which we could call-up, rather than having the operating systems on our own computer. We actually wouldn't even need a computer.

No wonder computer prices have dropped so low.

It looks like that is where technology has headed.

It's a dirty little secret, but for your average tech company, the annual Comdex conference has nothing to do with sales. Nothing to do, either, with advertising or getting on Web sites like this one.

Comdex is there so everyone can check each others' homework and make sure they are on the right track.

This year, it seems everyone is, if not on the right track, then at least on the same one.

Forgive the buzzwords, but the future according to Comdex 2000 is "broadband," "wireless," "distributed," "standards-based," and "OS-agnostic."

These baffling bits of jargon do make sense, but let's take them one at a time.

"Broadband" first: Internet membership is growing at an ever-increasing rate. In the United States, roughly half of the population uses it. It has become an essential business tool, and a personal necessity for millions of users.

But people are coming to find out that using the Net from a standard dial-up connection with a modem is too slow. Faster and faster connections — broadband, in other words — are needed, and the technologies that facilitate this were abundant at Comdex this year.

More importantly, a great deal of Comdex is given over to what we can do with such sizeable connections to the Internet. Voice-Over-IP (VoIP) technology is making a great show.

Using your computer and Internet connection to make ordinary telephone calls is proving to be just as good as a standard phone line, but for a minuscule cost.

Indeed, companies like Net2Phone already give completely free calls to anywhere in the world over the Internet. Expect to hear plenty about VoIP in the next twelve months.

Expect, too, to hear about the second buzzword of the show, "wireless." Wireless was king at Comdex2000. Vast swathes of the 1 million square feet of floor space at the show were given over to wireless in its various forms.

We have the latest in wireless networks for offices and the home that can connect computers together at very zesty speeds.

We have Bluetooth, the short range system for connecting lots of simple devices, like phones, PDAs and stereos together, that is finally beginning to appear, and we have Internet-capable mobile phones with all the flavors, shapes and novelty ringtones you can handle.

Add broadband and wireless together and you have a logical jump into the biggest trend for 2000: "distributed."

The advantage of wireless is that you can carry the device around with you. The problem then is that it has to be small enough to lift.

To make it light and small enough to take advantage of the wireless technology, you have to leave bits out and to make the devices really cheap. It's a good idea to leave out all the heavy expensive stuff — hard drives, for instance, and powerful processors.

You would think this might create a problem. Today's Microsoft Office software already requires more computing power than was used to power the first space shuttle, and leaving parts out ought to mean you couldn't run even the simplest word-processing package.

Not so. By having the big and complex programs running on a computer "somewhere in cyberspace," the user can work with them using only a simple Web browser. Get online, connect to the correct "Application Service Provider" (ASP) and Word, say, will appear inside your Web browser's window.

Web browsers are comparatively simple programs and require much less computing power, and so can work perfectly well on one of the new wireless devices. All that is required is a broadband Internet connection fast enough to make it feel as if the program is really running on your own machine.

There are still some snags to be ironed out when it comes to this way of doing things, but it does have plenty of advantages.

First, it is very easy to administer. All a user must do is connect to the correct Web page, and the program is there waiting for them. No more messy installations.

Second, the software should always be the most up-to-date, as it only needs to be running on one machine to be available for thousands of users.

Finally, it opens up a whole raft of new business models. Why spend $200 on a new program, when you could rent it for a few cents a day? Why bother buying a massively powerful PC when a small Web-device and a subscription to the best ASP could prove more useful?

It's a powerful vision, and one that Microsoft's boss, Bill Gates, has fully taken to heart. His company's so-called ".NET" strategy is based around this idea. He says he has "bet the company" on it.

At his keynote speech here, he unveiled a prototype Web Tablet device to do just this sort of thing. It should be available mid-2002.

There is, and probably always will be, a debate as to the form of the newer devices. Michael Dell, the CEO of Dell Computing, thinks Gate's tablet-style computer would be limited by not having a keyboard, but he does support the whole idea.

"The world is rapidly moving from fixed to mobile computing," he said during his keynote speech. "This will have a huge effect on our industry, and it will change the way people work."

Either way, putting complex programs onto the Web necessitates the next buzzword: standards-based. As the Internet itself, and the Web on top of it, are both based on certain free and public standard ways of communicating, the ASPs must ensure their programs also talk to the Web browsers in these, and only these, languages.

This is significant for two reasons.

First, the languages are freely available. Most book stores will sell you a guide to XML, for example. This levels the playing field somewhat: anyone can write applications for people to use over the Web.

Second, it makes the programs themselves subject to the final buzzword of this year: "OS-agnostic."

Today, software writers have to create a program specifically for each type of machine: PC owners cannot run Macintosh programs and so on.

With Web-based ASPs, all you need is a Web browser. It doesn't matter what system it's running on. As far as a Web-based version of Word is concerned, there is no difference between a browser on a PC and one on a PlayStation 2.

This makes a massive difference. Operating Systems like Linux, itself making a huge showing here this week, and BeOS are moving rapidly into this arena.

Linux, while still not ready for the average user's home machine, is rapidly taking over both the server and the embedded marketplaces — in other words, the big computers at one end and the little ones, powering cameras and refrigerators, at the other.

What we are looking at here is a fundamental change in the way a lot of computing is done. It depends, of course, on many technologies developing together — high-speed Internet and high-speed wireless specifically — and it might not happen for a while yet.

But with so many millions of dollars, and thousands of developers working towards it, we could be in for some very exciting times.



-- Anonymous, November 26, 2000

Answers

ASP is not a "done deal". Maybe if MSFT and IBM are the ASPs, there will be wide spread adaption of the "app online". But, what happens if you use a service bureau for your financials and taxes and they go Bye Bye? "Rent an app" has been around for a long time. Service bureaus made people like Ross Perot very well to do. The real question is "how much of your internals do you want outside your Company's walls?".

Wireless is a hype until a universal device appears that doesn't require a magnifying glass to see with and a tooth pick to type with. FIBER is "where its at". Check out the history of "Level 3" and see what they are getting ready for. ............. xxxxxxxxxxxx XML may seem easy enough but building full scale apps. out of it will require the ability to multi-task in Java, C/C++/C# and scripts, know some networking and most importantly, be able to think in data base. XML and XSL are only the surface of being able to put together the interfaces of true Value Chain "B2B Communications". If anything, the demands of XSLT and schema will force people to get their acts together. There will be NO "applets" of XML/XSL in E-Comm. No things "dashed off" by teenagers to jazz up Corp. Web sites. In addition, the Network and demands for security will increase. XML has one great thing going for it: CHEAP. XML/XSL is easy to understand if you have ever built a Data base app. or designed a major app. If not, you will be lost in a vast maze of existing legacy apps. The one great App. that will happen will be the extension of XML for EDI apps. by the 80% of the companies that could not justify the EDI costs before. ...................... It may be true that the PC market will slow. The need for the power of a 1987 era Mainframe on your desk is questionable. We need a new "killer app" to drive another wave of PC buyers. I can add ram to my PII and PIIIs and as long as I have big pipes I probably don't need a next gen. 1.5 gig with 80 gig HD and a Gig of ram ............YET. However, everytime I think that, something major comes along and we are off to the races again. I think the next driver for this form of insanity will probably be the arrival of the always on, big pipe which will deliver everything to "outlets". Then by 2005, you are probably looking at on demand movies etc. Who knows? We can't manage what we have now. Upgrading "people skills" even at the highest level is almost impossible. At MSFT you have to get W2000 certified and many have not gone to NT yet.

-- Anonymous, November 26, 2000


Real question is security. How much do you trust those off-site apps?

Industrial espionage is a multi-billion dollar business, and even one programmer in the right spot could open up all sorts of mischief.

Security will be the limiting factor in outside apps.

-- Anonymous, November 27, 2000


I'm trying to figure out why anyone would "go to NT".

We're talking about an OS (touted for its alleged "stability") that's up to Service Pack 6.

And the company I work for runs mission-critical apps on it. But we're currently running SP4.

Shoot me.

(I've heard rumblings of UNIX/Linux, not to mention a return to Big Iron. And we've been in the process of migrating several of the more important databases to Oracle 8i. At least someone at the top of this food chain appears to have a brain.)

Sorry, but at this point in time Charlie, MS should stick to the "home-user" market. When, and IF, they can get it together and release a quality product that doesn't require 6 service packs to get it "right", then they can enter the world of Real Business.

-- Anonymous, November 27, 2000


I don't think you understand that MSFT is far beyond a software manufacturer for single boxes. They outgrew that by 1995. I also don't think you know how high the level of QA/QC at MSFT is now. The reason they still produce for single boxes is simple. There is a 10- 20 Billion dollar market out there in home apps. Games are estimated at 6 to 7 billion alone. CBT and CD Tutorials/education is growing very fast because they can be supplemented "online" but people still like the CD and "a book" in front of them.

However, the vast bulk of MS's revs. come from bus. and bus related software. And the need for stability and security mandates MS software should be on an MS OS. Exchange 2000 starts at about $4,000 and you have to sell 60-80 copies of Win ME (dog). Exchange 2k needs Win2000 to run best and implement its functionality. ETC.

MS's future in in the business market not single boxes. They have known that since they started the NT core group in 1988. The head of that group 12 yrs. ago came from DEC and DEC is no more. MOST problems with NT come from app. software written for W95/8 which is a different kernel. I have to use a piece of garbage RE software that is nothing but a series of patches from DOS to Win 3.1. Most people can't get it to work under Windows2000 but I found a way. It leads to all sorts of problems so I off loaded it to an older PII and now my PII /W2k NEVER crashes........NEVER. The older box is still running a BETA of W2k good to Feb. So I live with the problems.

NOTE: it was the "app"....NOT NT or W2k.

In addition, the nature of ANY OS is that it will contain "conflicts" that only surface in the field. 85% of ALL problems with Windows are traceable to: other manufacturers' DRIVERS. Of the balance most are simply : "RTFM" problems caused by users.

When W2000 was in R&D, the rules were laid down that all app. software would comply to W2k to be certified. And, the Blue Screen of Death was almost eliminated in W2k. When a conflict arises, the app. freezes and can be stopped and restarted instead of the entire OS. That is a major change and that ALONE is worth the upgrade from NT. Most of the reports of bugs and security holes are coming from MSFT itself. MS finds it cheaper to do that and fix immediately than to try to use "customer support" on a case by case basis. Their support costs now are dropping because the QA/QC is showing.

Service Packs mean nothing. They could have simply sold NT 4.0, 4.1, 4.2 etc. As for "bugs". All software of any substance is released with bugs. Even the "holes" reported in NT/W2k are minor when compared to the scope of the programs involved. Hackers are constantly probing for those holes and most simply report to MSFT which issues the "alert notices".

I was a big user of dBASE II, III and IV. Six months after IV was released, Ashton Tate (later Borland) announced that they had over "500 anamolies" (PC for bugs). Most were trivial and could be solved with "work arounds". That is what you see with NT or W9X or W2000.

Most are minor fixes of things like drivers that conflict. They replace "patches" and hot fixes. In the future, internet delivery of such things will obviate the need for the service packs.

FYI: Windows2000 is a far better product. It is the product of over $5 BILLION DOLLARS/YEAR IN MSFT R&D.

That is slightly behind only IBM in R&D. But MSFT's efforts are almost totally dedicated to Software whereas IBM's extends to fundamental research in diverse areas including quantum physics and solid state physics.

Point is $5 BILLION IN R&D......is not a joke. It is that expenditure that you are only beginning to see with the MS O/S, Apps and even I.E.

-- Anonymous, November 27, 2000


I also don't think you know how high the level of QA/QC at MSFT is now.

Considering the absolute garbage they release, I don't think ANYONE knows this.

And the need for stability and security mandates MS software should be on an MS OS.

And you'd think they'd want to do something about this, wouldn't you? But you see, Charlie, They Are Microsoft and don't have to do anything about it. People like you will continue to buy and tout their products ad nauseum, regardless of the fact that they are unstable.

MS's future in in the business market not single boxes. They have known that since they started the NT core group in 1988.

And it's been twelve years and the OS is still a POC. Wow. Let's hear it for $5 BILLION IN R&D.

BTW, apparently YOU are unaware that while the majority of the problems come from other companies' software, the reason there ARE so many problems is due to MS's .dlls.

And, the Blue Screen of Death was almost eliminated in W2k.

That would be the Blue Screen of Death that was supposed to have been eliminated with Win 98, right?

Look, "tout" all you want; MS is not in the same league as the Big Boys ..... yet. Whether or not they want to play Real Business is a decision that can only come from them. Let's see what >$5 BILLION IN R&D will do.

-- Anonymous, November 27, 2000



You have to remember that I have been an IBM fan for only 30 plus years but by going to all the MS functions and reading a ton of white papers on the ms web site, I found out by 1998 that MS was no longer some bunch of former teenage game players but a serious player in Corp. Land. I don't get into religious wars about Unix vs. Win or Oracle vs. DB2UDB. Some people still swear Atari made the best PC ever. Who cares?

The fall of companies like MarchFirst(WhittmanHarte) and decline of Informix should warn anyone that "past successes is no basis for future profits". Only the powerful balance sheet and strong cash position of Novell gives the N-True Believers hope that they will "live to fight another day". REALITY is IBM and MSFT and alliances on both of their sides. In fact, take a look at what IBM Global is doing with Windows2000 and SQL2000. Then ask, why would IBM Global be selling MS products when they have their own? The answer is part of Gerstner's strategy and he must be doing something right if IBM Global has gone from 2 billion to 15 Bill. in sales in the last 3 years. Now, if Windows 2000 is GOOD ENOUGH FOR IBM....what is the problem??

Wrong. On blue screen. The blue screen of death is the screen for a freeze from NT. I REPEAT: W2000 properly configured does not freeze. PERIOD. And I run upgrades on all app software and browsers.(Except I'm passing on the thrill of Netscape 6.0 after multiple disasters with the betas.)

Running them and betas is an invitation to system problems which JUST DO NOT HAPPEN. AND......as a bonus, OFFICE 2000 NEVER crashes. There is now a deal in Corporate land for Open Licenses, Office Premium for the price of Office Pro. That is like a free upgrade for the O-97 trailing edgers. Not to mention that O2k will do everything that Lotus Smart Suite Mille. Ed. AND......Notes/Domino will do. That leaves the two (IBM and MSFT) as the only cost effective TRUE Collab. Solutions on the market for SMEs. Toss in the 'one hour intranets' that users can set up for Work Group Collab. and you have another example of **results** from the $5 billion R&D.

Wrong again on "the big boys".

MS is in the league of the "big boys". Five years from now only MSFT and IBM will be standing of the current vendors. Should MSFT enter the Unix/Linus(x) world, you can shorten that to *three* years from now.

One thing is for sure. They will not release an OS for servers with a known "memory management" problem that screws up anything placed on the servers "randomly" so that you can't get a handle on fixing it.

-- Anonymous, November 27, 2000


Charlie, I'm not debating the "merits" of O2k; I'm talking about the $5 BILLION IN R&D you're claiming on the part of MS. IF (and that is a conditional IF) MS is in fact allotting that much money, into where are they researching? Into what are they developing? What are they doing about all the CURRENT customers that they have seemingly "left in the lurch" UNLESS said customers upgrade to the latest and greatest from Whiz-Bang Land (at the appropriate fee, of course)?

Don't you see this is their "business strategy"? And whether or not "everybody does it" (vis a vis releasing buggy software) is NOT a defense. It is a practice that is INDEFENSIBLE. And yet we, the ever-gullible, ever-wanting-the-latest-gadgets, naive buying public continue to fall for it. Shame on us.

Charlie, my feeling is that businesses do not want to invest all that much in anything MS (the thrill is definitely gone and I think old Bill knows this), least of all mission-critical applications. MS has a fairly shattered "business" reputation. I suggest to you that they need to upgrade THAT before they can go further. If that's where the $5 BILLION IN R&D is going, well, it's about freaking time. (I would also suggest to you that any "partnership" between IBM and MS was NOT the brainchild of Gerstner, regardless of what the "press release" stated.)

As to the blue screen of death, sorry, but you are wrong. I run Win 98 at home (on two PCs) and I get the blue screen of death ***RUNNING MS PRODUCTS***, nevermind any (GASP!) non-MS software (which is a .dll problem.....every single time). (Note that it does not happen on my little dinosaur of a laptop that is still running Win 95 -- with ANY software.) And when Win 98 was released, its "selling point" was that it eliminated the Blue Screen of Death. Oops. Guess none of that $5 BILLION IN R&D made it into Win 98. (And what is that up to now -- SP2? SP3?)

I don't really have an opinion on this one way or the other, but my gut feeling tells me the day you see MS going the way of UNIX/Linux, is the day MS admits defeat. IOW, I don't think you're going to see that anytime soon.

Bottom line is that I don't care what I use, as long as it works. And my opinion (and that of many colleagues over the years) is that MS just doesn't work ..... in business.

Yet.

-- Anonymous, November 27, 2000


I don't think you understand that MSFT is far beyond a software manufacturer for single boxes. They outgrew that by 1995. I also don't think you know how high the level of QA/QC at MSFT is now. The reason they still produce for single boxes is simple. There is a 10- 20 Billion dollar market out there in home apps. Games are estimated at 6 to 7 billion alone. CBT and CD Tutorials/education is growing very fast because they can be supplemented "online" but people still like the CD and "a book" in front of them. Basically this is what I was trying to get across. The home consumer does not need toi buy a 2-3K computer with applications they never use, much less have any clue what to do with.

This is where it is splitting up. Business and individual use. Get a $200.00 "box" and your teenager can watch music videos and surf the web and have a little tiny bit of power to play games. Such a waste of good computer ability will no longer be wasted on brain dead "home computer users". Anything more complicated they may want to try they will find on-line on a bigger "box".

Boxes can be made and sold which people can customize with their preferences.

Big Boxes are already coming back. Sucessful on-line businesses have to go back to mainframes because the smaller ones are incapable of handling their needs. Ebay is an excellent example.

-- Anonymous, November 27, 2000


http://research.microsoft.com/research/

LINK TO MSFT RESEARCH PROJECTS LISTS

-- Anonymous, November 27, 2000


CHERRI,

**DEFINE BOX**. Even if you tie all the boxes together and make "the network is the computer", you still have to cut out smaller boxes as an "interface". Proprietary concerns suggest that you will not live long enough to see all business apps. on "one big box". "Cooperation" with competitors in a Value Chain ENDS at the Sales dept.

A Server Farm can adequately handle most of the needs of even the largest companies except for things like billing phone calls by the seconds;. That requires massive transaction power. However, DELL runs its entire B2B and E-commerce app on a farm of Dell Servers and Windows 2000 with a Main Frame Backend to its ERP system. Dell's app. is the largest in terms of dollar vol. of all web site ventures. I think they have passed $50 million/day in orders.

One of the largest apps going sits on a server farm : MSFT's Terraserver.

-- Anonymous, November 27, 2000



OFF OFF

-- Anonymous, November 27, 2000

Charles,

They (ebay) need to be able to handle a lot raw data in the same place. People come in and search for things using their own unique set of attributes. They have to be able to merge and sort beyond what the little boxes, even tied together can give do for them. When you have a half a million people trying to do that at the same time you need one big box, the little boxes just aren't able to handle it, and they are loosing sellers and buyers over it. Not all customers have the same needs. Some require a technological step back to work efficiently. I may have found my little ninch *grin*. I am right about the home boxes, they will be so cheap that they will be issued like books in schools pretty soon.

-- Anonymous, November 30, 2000


Moderation questions? read the FAQ