Microsoft Searches for Search Answers

Microsoft has more or less conceded Internet search to Google; there's no force powerful enough -- not Yahoo and not MSN Live Search -- that can topple the Google. But enterprise search is another matter, and here Microsoft may at least have a shot, if not an edge.

That's exactly why Redmond bought FAST, a Norwegian search company. It seems that Microsoft will tie the FAST search engine to key server apps, such as SharePoint. This is an open system allowing SharePoint developers to include custom search in their apps.

Mainframes Make News, Episode 1
I cut my journalistic teeth on MIS and mainframes. As a young buck writing for Computerworld in 1985, it was all CICS, terminal emulation, spaghetti code and leased big iron. A year or two later, pundits were predicting the death of the mainframe, upon which IBM exercised a Microsoft-esque (big) iron grip.

Twenty-four years later, I'm more of an old sow than a young buck, but the mainframe is as young as ever -- and so are lawsuits over IBM's monopoly. The old mainframe cloners like Amdahl are long gone, turning a mere monopoly into a sheer monopoly. And IBM will do whatever it takes to preserve its market grip.

Platform Solutions built a tool that let commodity servers run mainframe software, and IBM wasn't too happy. It tried to stop Platform, then bought the company so it could shelve the technology that could've changed the fundamental economics of mainframe computing.

Many in the industry are complaining, including a trade organization that counts fellow monopolists Google and Microsoft as members. My guess: IBM's mainframe grip will remain as strong as Google's and Microsoft's.

Mainframes Make News, Episode 2
There are two reasons mainframes have survived: They handle big apps very well, and there's little benefit to redoing much of this software to run on other architectures. And mainframes, if configured right, are greener than a Tiger Woods' dream. The highest-end IBM System z mainframe, for instance, can act as 1,500 separate servers.

I looked into the System z and while it's designed to run Linux VMs, I was told by an expert I trust that there was no architectural reason it couldn't run Windows, as well.

Well, one ISV apparently found a way to make it do so. Mantissa Corp.'s z/VM tool is being prepped, and if it works large shops may be able to save massive amounts of electricity by consolidating Windows servers onto mainframes. That should make two monopolies, IBM and Microsoft, very happy. The electric companies may be less pleased.

Do you still care about mainframes? Big answers to this big iron can be submitted to [email protected].

Your Turn: IT Gone Good
Two-and-a-half years ago, I wrote a story about IT abusing its power -- blackmailing executives, spying, stealing and sexually harassing.

I'd love to do the opposite, to show where IT uses its power for good. Do you volunteer and use your skills for good? Does your organization itself do good and have IT systems to support those efforts? If so, tell me your tale at [email protected].

Mailbag: Thoughts on Vista Capable, IE 8, Windows 7, More
Marc thinks the Vista Capable lawsuit is still much ado over nothing:

I am surprised we are still talking about this! The Vista Capable specs (800MHz, 512MB of RAM) were no more ridiculous than the original Windows 2000/XP specifications (300MHz, 64MB of RAM). Microsoft has a long history of stating MINIMUM specifications which were technically accurate, but deplorable just the same. Should someone have raised a stink in 2000? Probably, but nobody did. Were the Vista Capable specifications misleading? Well, yes! Practically, though, what was the extent of their damages? If you bought a brand-new low-end machine in January 2007, it was equipped with a 3GHz+ Celeron, 512MB of RAM and integrated Aero-capable graphics capability -- and you paid under $400 for the system. At that time, 512MB of RAM cost about $50, so the REAL damages for a buyer of one of those systems was the cost of a RAM upgrade.

If anyone deserves damages regarding the Vista Capable moniker, it was those folks who went out and bought shrink-wrapped Vista code expecting to install Vista on their (circa 2000) 800MHz, 512MB machine. They paid anywhere from $100 to $400 for Vista expecting it to work on LAME hardware. No amount of hardware upgrades would make such a system "acceptable."

IE 8 isn't perfect yet, as these readers point out:

I was glad to see the Compatibility View feature in IE 8 but rather quickly found it doesn't work. I have had numerous sites that haven't worked with IE 8 and have tried CV, which still failed to work. The biggest is my Sovereign Banking account which is more or less due to a bug with the Web site, but Compatibility View should still fix it. I get triple-prompted for my log-in on that site with IE 8, which makes viewing my account online with IE 8 a huge security hole.

I installed IE 8 without a problem on Windows XP SP3. However, it wouldn't run. I kept getting a memory error. I searched the Internet for a solution, but couldn't find one. I then opened up the Control Panel and clicked on Internet Options and went through the tabs. I went through the list in the Advanced tab, and when I came to "Enable memory protection to help mitigate online attacks," I unchecked it, clicked Apply, then OK. I retried IE 8 and this time it worked.

I still haven't figured out why having that checked caused a problem.

Meanwhile, Windows 7 is still looking promising:

I've been using the Windows 7 beta on a personal notebook, and I am very pleased with it. It's more stable than Vista, seems to be faster and some apps that will run on XP but not on Vista seem to work fine on Windows 7. Hardware compatibility is at least as good as Vista. I have found that MS Office applications frequently die for no good reason in Vista, but not in 7. Nothing I have tried has been less reliable in 7 than in Vista, and several things have run better. At this point, I would not advise anyone to buy a new Vista PC or a Vista upgrade; I would suggest that they wait for Windows 7.

I am so impressed with Windows 7 that I have loaded it on most everything I can. Being that this is still beta code, I am wondering what the final release will bring. Even as just an end user, I find very few compatibility issues, no lock-ups, and I even see that disk defrag runs on its own schedule! From what I have seen so far, Windows 7 even works better than OS X Leopard. Hell, I am even running Win 7 on my MacBook!

Speaking of Windows 7, Gartner recently sent out some mixed signals about the new OS, prompting Doug to ask readers if they trust the opinions of IT research firms. The answer was a resounding "no":

Neither use nor trust them. I believe they are bought and paid for by vendors.

Every good analyst should know that if you pay the Gartner Group enough, it will say what you want it to. That is what our company found out, anyhow. I would rather read your column and other blogs than trust what they say.

IT research firms are essentially useless. I have never seen a good example of their research providing useful information, but I have seen several examples of their work being used to support the decision someone has already made. If you look, you can find "research" to support pretty much any position.

I have read many of Gartner's pronouncements over the years, and they seem to fall into one or another of these categories: 1) so obvious that nobody will actually learn anything, 2) essentially meaningless because the writer has hedged so much, and 3) just plain wrong. This is not limited to IT, but I think IT research firms are especially prone to it.

Maybe I have become jaded in my not-so-old age, but I have come to trust research firms like used car dealers. Avoid them if at all possible, but if not, take everything they say with a grain of salt.

Every time I have asked for one-on-one time with an analyst, I usually get a "handler" and an analyst. I believe the handler reports if the analyst doesn't mention at least three companies (clients, of course) that provide services in the area that you are looking at. Research firms are businesses; they live and die by the revenue they generate. I am not saying that they would be partial, but like with print media, subscriptions alone cannot pay the bills.

Tyler shares his thoughts about why the press can sometimes, well, stink:

I blame the degradation of quality within the media on shorter and shorter cycles with editors simply not doing their job. There was a time when editors rode herd on their writers and insisted on quality, second sourcing, fact checking, etc. Now they're just as desperate as the writers to get the story out and pray for a few readers.

Which leads me to the second reason quality is in the crapper: Sturgeon's Law.

But one thing that hasn't degraded in quality is Doug's vivid similes, with which regular readers -- and Jeff -- are quite familiar:

How do you come up with these comparisons that are more colorful than a psychedelic dream sequence?

Watch this space for more reader letters next week! In the meantime, share your own thoughts by commenting below or sending an e-mail to [email protected].

About the Author

Doug Barney is editor in chief of Redmond magazine and the VP, editorial director of Redmond Media Group.


comments powered by Disqus

Subscribe on YouTube