Blog This!   Lee Geistlinger's Web Log

Blog Home
Blog Archives
LittleGhost Home

E-mail: geistlinger AT gmail.com

Everyone loves lists Books, Movies and so on

My Online Aggregator

OK, my last two blogs pretty much discussed the "birth" of the computer generation and then dissected its current growing pains.

While it might be difficult for industry members to agree on what changes are currently underway, I don't think you'll find many who think that things are not changing dramatically.

Which begs the question, What Next?

What indeed.

While I don't think the future of the IS/Internet industry is black and white -- say, for example, Microsoft will rule all or be crushed beyond recognition -- I do think the divide that exists today between beheomoths such as Microsoft and smaller sites like salon.com will continue to grow.

We are heading for a larger division between the haves and have nots. This will force more bankruptcies and more consolidation/amalgamation. As I mentioned in my last blog, this will leave us with fewer cool sites while at the same time giving us more -- for example -- real-time news channels.

There are other upsides and downsides of this increasingly large division, but that's a good example.

So what else does this gulf mean?

• More and more sites will be brick-and-mortor based. For example, Kohl's department store just now opened a Web site. They rode out the Web/not_to_Web storm sans site, possibly learned from other's errors, and now they are on the Web. Purchases there can be returned to physical store.
  
• Leading from above, there will be more integration of the physical and virtual stores.
  
• This does not mean that pure Internet plays, such as Amazon, will not come up/surprise/endure/drive other Web endeavors.
  
  
  
  • There is always room for niche players, for one.
    
  • Second, there will be a lot of so-called pure Internet plays that are actually backed not by VCs but by brick-and-mortar companies. Example: Orbitz. Has high profile already (heavy TV ads), exists only on the Web...but is actually a product of the joined forces of several major airlines. (Designed to combat the other online reservation sites.)
    
  • Third, there is always room in the Internet space for another Amazon-type company: A company leveraging the Internet to do what the Internet does best (distributed; low human-interaction; always on; repetative and redundant). I see the potential for one of the many online storage sites to go enterprise and become a Fortune 500 backup solution. It will happen, and only a handful of companies can support this type of infrasturcture (IBM a player?).
    
  
  
• Big Winners: Established, well-entrenched companies. Look at it this way -- one of the reasons for the Y2K scare was because of business' COBOL programs. Who programs in COBOL now? No one. But businesses have a slew of legacy COBOL programs (accounting etc); still do. While COBOL -- from the 1970s, I think -- is NOT being used except for maintenaince now, it's still here. Same for all other well-entrenched systems. Yes, Microsoft upped its licensing fees. You have a $2 million MS installation. You won't toss that to avoid an extra few thousand in fees. No way. For the same reason, Linux will never really get to the desktop until programs that can read/write MS Office exist. You HAVE to have Word and Excel in business. Period. Need Windows or Mac to run it. So Microsoft will do well in the near future, as will Oracle, Sun, SAP, PeopleSoft and so on. Installed base is huge. Think of it this way: People are VERY reluctant to changing browsers. Think how much easier it is to change browsers than a database installation/apps written against it. Not going to happen quickly, at least.
  
• Big Losers: Those without installed bases. Small, cool applications or technologies. They will have some legs, and some may well survive and do a fair amount of business. But for the most part, things outside the mainstream -- now that the Internet IS mainstream -- will either wither and die or become, like reading webmonkey.com, inconsequential. This, unfortunately, means most open-source products/projects. Don't believe me? Take open-source databases as an example. All tests and laundry lists of features demonstrate that PostgreSQL is far superior to mySQL. The latter has a wider installed base; I have yet to run across anyone (outside of on the Net) who uses this "better" database. Installed base. No one is going to change, even in this open-source environment. RedHat just picked up PostgreSQL as their bundled database solution (to battle Oracle for smaller companies); this may work, but they still push mySQL over PostgreSQL for average use. Go figger.
  

Let's get a bit more specific:

Winners:

• Microsoft -- Even if the company is broken up (I doubt it) and Linux takes over the desktop (yeah, that'll happen...), they still have Office, which is the de facto standard of the computer world. And today, the computer world = the business world. May have lower profits, but will still have enormous profits. MS is also very Internet focused. Be afraid. Be very afraid. On the other hand, MS has done -- legally or otherwise -- a good job of driving standards (MS products..). That's a Web thing. That's good. Whether or not it is a good thing that MS is doing it is another thing, but isn't it nice to send an Excel spreadsheet to whomever and not have to worry if they can view it? OF COURSE they can: MS (and it's products) are de facto standards. Good. And scary...
  
• ASP (Microsoft) -- Sites will become more and more dynamic in the near future. Requires both database and scripting tools to deliver this stored data. The scripting tools of choice will come down to two, I believe: Java-based tools (Broadvision, WebLogic) and MS ASP. That pretty much means a good hunk of the pie will go to Microsoft, and it makes sense. Java runs best on UNIX platforms; UNIX platforms are more expensive, Java is slow to deploy and "Java" (Sun) doesn't make a database. Have to hit Oracle or IBM. With MS ASP, one platform for server and database. And guess what? It's the same platform that the office uses for Word and Excel. Keeps network admin costs down. And ASP runs on cheap Intel boxes (as does Linux; another story but noted). Compelling. And while I am not sold on the whole ASP/COM/DCOM model, I don't know enough about it to really make quality comments. Put it this way: It's a very compelling solution to the non-tech business managers. No NT and UNIX admins; one set of licenses, one kind of boxes and so on. From a company they've heard of. (VERY compelling.) And ASP is fast -- it's all native, all the API hooks are there. The one caveat is security: ASP is really a Windows-only solution, and IIS is not a secure Web server. Yes, can put iPlanet on Windows, but that kills some of the native speed advantages and other things like that. Not compelling. MS had -- deservedly so, I believe -- for security issues with IIS lately; they had better shape up with this or quite a bit will be at risk for Microsoft personally.
  
• Oracle -- Yes, they have been hurting. But that's relative. They have just been doing LESS well than before. Still quite fine, thank you. The business world is going more and more Web oriented. Web sites are becoming more and more database dependent (static sites are a thing of the past, and will be anacronisms in the near future). Oracle is the top-end database for business. Do the math. Unless IBM gets really serious with DB2, Oracle will continue to rule. I don't see much of a threat from Red Hat's database (PostgreSQL): That's more of a solution to a problem most of those businesses didn't really realize they had.
  
• Perl -- While no one except consultants and O'Reilly publishing makes money from Perl, it will still endure. Its role has changed, from the duct tape that holds the Internet together (Larry Wall) to the pipe that transforms legacy data to current containers (databases), but it is still ubiquitious, fast and unbelievably useful.
  
• Apache -- This will survive because it has a large installed base and because IT IS FREE. One of the may neat things Netcraft has done is to give a better picture of the servers operating on the Net. Yes, Apache rules. Yet the biggest sites -- the ones that you've heard of -- run something other than Apache. Usually IIS or Netscape. They are more set up for the dynamic, e-commerce type of Web that we have today. I think Apache will survive -- one can do a lot with it -- but it will always be marginal in this regard. While it may power the
  "majority" of the Internet, that will mean my site and "ilikecrayons.com" -- in terms of page views, I believe non-Apache sites already win or will shortly. Apache will NOT be the choice of CNN, Amazon etc. Universities will use it; so will personal users. Ask yourself this: With all the publicity -- virtually none negative -- and support Apache has, how come v2 still isn't here? Been a LOOONNNGG time not coming.......
  
• Unknown publishing tool -- While there will always be a need -- increasingly significant -- of people who actually know what they are doing on the Web, the introduction of good publishing tools will allow those who don't know and probably don't care about TCP/IP, sockets, CSS and so on to publish good material. Did editors/writers in the near past have the knowledge/desire to run printing presses? NO! So why should the same group have to learn "stupid browser tricks" to get the article about fish food on the Web? Vignette promises a solution; it's not there yet. Allaire's Spectra is a valient attempt; I don't see it working. To date, there have been a lot of bad, high-profile Java solutions out there. ADG Dynamo's is the best; still not there yet. This is hard shit, and is very business-dependent.
  
• Unknown killer app -- The Web needs a killer app. Yes, had HTML (which defines the Web, not its killer app) and Flash (mainly evil [used improperly] ). XML is not a killer app, it's an extension. Need that fresh blood. Could be a new publishing system that actually works.
  
• Java -- Yes, check your prejudices at the door. I just read a report of some report (how official..) that said that in either 2002 or 2003 Java programmers will outnumber MS programmers. C# programmers are not even visible. Java is good; sure, it is flawed, but it can be fixed. And for enterprise projects, has no peer except ASP, which is powered by .... MS programmers ......
  
• Database-driven tools/technologies -- Yes, XML is there. That's a given. So is JSP, ASP, CF (to a very moderated degree), PHP (to an almost invisible degree) and so on. This is the hook for the publishing tool. Whichever KILLER publishing tool comes out, its database will both be part of its success and part of the future success of the database.
  

Losers:

• Consultants/consulting companies -- Back in the WWW heyday -- before March/April 2000 -- consultants ruled. Why? Because businesses had desire to get on the Web and no clue as to how to do so (the COBOL or FoxPro pros in IT has no clue what HTML is...). Consultants did. Businesses had money, and -- and this is key -- consultants had the ability to quickly (relatively....) get a business online at a time when speed was on the essence. Many pluses to being first. Today that is not true at all. And DigitalWorks is almost dead, Xpedior is dead, Scient has merged with iXL (both companies hurting), Razorfish is bleeding...marchFirst/Whitman-Hart/Divine (hahahaha...). Consultants will survive, but in a greatly modified, highly targeted, lower pay way.
  
• Open-source technologies -- Yes, that's a bitter pill to swallow, but with few exceptions (Linux, Apache and Perl) this is the handwriting on the wall. One of the reasons MS has such a large number of developers is that they make such good tools (Visual Basic, Visual Studio etc.). Open source is, right now, focused on the technology, and there are, for the most part, no tools. PostgreSQL, a great, fast, stable database, has one tool -- a Linux only (! who puts a REAL production database on a Linux box with a GUI installed???) piece of shit called PGACCESS. I never use it; I use command line. MS SQL server's tools rock. I'm on v7 and there are gaps in the tools, but way way way way more functionality, more intuitive than PGACCESS. PGACCESS looks like something I wrote, and that's not a good thing (very unstable, as well, another hallmark of geistlingerWare). Let's examine:
  
  
  
  • PHP -- Great language that combines the text-friendly tools (regular expressions) of Perl with the power of UNIX (*NIX, if you want to be a purist) and maintains a very nice balance between combining the scripting language capabilities of Cold Fusion with the N-tier approach of JSP/servlets et al (separate functionality and appearance). Very popular with adult sites and company sites (i.e. small businesses, such as "Hal's House of Records") -- the latter programmed by a single person who is a geek. Reason for popularity: FREE. Who would pay for this or the Zend engine? No one. Reason for Death: Lack of tools; lack of support (yes, the geek programmed it and it sings. Guy goes to college...who can maintain it? Go with ASP or Cold Fusion).
    
  • PostgreSQL and mySQL -- They will survive -- mainly mySQL -- to power smaller sites, but one of two fates face them: 1) They offer few tools and improvements in the near future, and they become marginalized. 2) They offer more/better tools and improvements in the near future, and they are forced to charge for them. And they become marginalized.
    
  • Napster/GNUtella/Bear Share et al -- Good tools; they draw the attention of corporate lawyers. Yes, they are screwed. They will endure, but not triumph. They will be eclipsed by software commissioned by corporations. This is good and bad, but get used to it. It is also the future.
    
  • Remember -- just because it's open-source does NOT mean that it's evil/excellent (depending on your politics). Cool/slick software is cool and/or slick, but that does not make it necessary, compelling and so on.
    
  
  
• Cold Fusion -- Cold Fusion is the single best language to quickly get a dynamic (database-driven) site up and running with a very low learning curve. If you know HTML, learn some new tags and some SQL and you have a dynamic site. How cool is that? That's the upside. Lots of downsides:
  • Since it is so easy to learn, the ranks of so-called Cold Fusion programmers is heavily littered with those who 1) Really don't know programming, 2) Really don't know SQL. Result? Inefficient pages; slower than it should etc.
  • Companies deploy it on stupid platforms (example: Access and Win98) BECAUSE THEY CAN. When at SOS, I think we had a great solution: CF on Linux (FAST!!!); MS SQL v7 database on NT. Yes, split platforms. Intentionally. Rocked.
  The bottom line is that CF will endure but will, like PHP but in a different way, be marginalized. Still a very good platform, however, and I've done both.
  
• Generalist Personal -- Like me. Fossils. Will still be valuable to those that have (accidently) hired them; VERY few will be hired for their breadth (with lack of depth). Yes, I'm personally screwed. I have to live with that....
  

While this current State of the Internet/Web is not generous to me, it is realistic, I believe. Again, that's not good for me, so that's a somewhat compelling reason to believe me.

Regardless of what I've outlined above, things are changing. Whether or not they change the way I've outline or not is ... well, history will judge (like History will every get its sorry ass over to this page!). This is what I firmly believe and probably have not set down clearly today; I'm certain that at least large chunks of what "I believe" in this entry will seem silly in the very near future.

That's me.

Unmasked.

Let the future come. It'll be interesting.