AOL is switching to Linux for the same reason most large companies make the change: to save money. Thousands of AOL servers are already 100% Linux, and more are switching over every day. AOL number-crunchers figure they can replace an $80,000 box running proprietary UNIX with two $5,000 Linux boxes and get a 50% increase in performance in addition to the cost savings. "Don't tell our competitors," one of our AOL contacts says. "Let them keep buying expensive crap."
We hear that every hardware vendor who approaches AOL is now being asked, "How is your support for Linux?" before they are even allowed to make a sales presentation.
Microsoft's server products have never been seriously considered by AOL, according to our insiders. "The licenses cost too much, their hardware requirements are excessive, they take too much labor to maintain, and we have enough security problems of our own without adding Microsoft's," says an AOL bean-counter who has access to the company's server cost numbers.
Good-bye Explorer, hello Mozilla
The Gecko rendering engine at the heart of the Mozilla Web browser is scheduled to replace Microsoft's Internet Explorer as AOL's default browser -- the one in the millions of free AOL CDs distributed every year -- in the 8.0 version of AOL's client software. (The current version is 7.0.) The Gecko rendering engine is already being shipped as a "beta" test product in some CompuServe client software packages, and reports from CompuServe users who have chosen to use Gecko instead of Explorer have been described as "very positive." This customer feedback is an important part of AOL's browser decision process. "We hear the question, 'What is the member impact?' whenever we are faced with a technical decision," says one of our contacts. And so far, it sounds like member impact of an AOL switch from Explorer to Gecko will be almost entirely positive.
"With Gecko, we have control over the client software and don't have to worry about Microsoft screwing up our streaming [audio and video]," says one AOL sysadmin. There is also concern at AOL about Explorer's "poor use" of the HTTP 1.1 Protocol. Our AOL sysadmin says, "HTTP 1.1 has lots more features than most people use," but AOL can make good use of many lesser-known ones like chunking, that are not supported by Explorer because, says our AOL sysadmin friend, "MSIE doesn't follow the spec correctly."
Even if future versions of Explorer manage to incorporate chunking and other features AOL wants members to use -- because they minimize download time and bandwith used per Web page delivered -- another AOL techie says, "It's still easier to optimize eveything when we finally control both the server and the client, and can make them work as smoothly together as possible."
All AOL tech people we spoke to denied that corporate dislike of Microsoft played any part in their preference for either Linux or Mozilla's Gecko rendering engine. They said their choices were made purely on what worked best in tests they had run; that their concern was not corporate politics but to make life easier and smoother -- and downloads faster -- for AOL members.
The only thing that might delay -- not stop, just delay -- AOL's change from Explorer to a Mozilla-based browser is allowing time for some of AOL's largest and most important "partner sites" to do away with any Explorer-specific features they have been using in place of W3C standards.
A browser shift by AOL is going to leave an awful lot of companies that assume their Web sites only need to work with Explorer scrambling to rewrite their code so that they don't lose AOL's 30 million-plus subscribers, or about 30% of all U.S. Internet users.
AOL for Linux users? Don't hold your breath
The basic problem with Linux support, says one of our AOL insiders, "is that AOL ALWAYS provides support for free. Hence the client is rather primitive/conservative in its feature set. This makes the AOL client reliable (relative to the software industry standards), because every 800-number support call comes right out of our profits. There are 15,000 AOL employees. Roughly 10,000 work at the Call Centers. We really, really don't want more phone calls from members.
"Now think of a Linux client. Either we completely disavow support for it (which is a very un-AOL thing to do), or we try to support every reasonably-up-to-date Linux config in the world. Even with the reasonably-up-to-date caveat, that is a hard thing to do. Where is the market and the demand?"
There was once a Linux-based AOL client "pseudo-computer" on the market that generated very few support calls, but that was because hardly anyone bought it. It was one of those "Internet appliances" every computer company was hot to sell a couple of years ago, but no consumers seemed to want it in place of a "real" computer.
Perhaps there will be an "AOL-compatible" Linux computer on the market one day, but chances are that it will be sold and supported by a company like OEone, Lycoris or even Lindows, who would probably just try to run the AOL client for Windows under WINE, anyway.
But don't hold your breath. No AOL employee we have talked to, at any level, claims knowledge of any current or future plans to offer AOL client software for Linux users.
What it all means
Obviously, a major AOL support contract would be a big win for Red Hat. It's not in the bag yet; negotiations are not complete and are still "very touchy," says one Red Hat person, and that's why Red Hat is still keeping mum instead of shouting joyfully from the rooftops.
If AOL's techies have their way, the contract will go through without further delay. One of them seems to think it is already a done deal, with only a little i-dotting and t-crossing left before it becomes final. "We get to bitch to Alan Cox about kernel problems now," he says exultantly.
On the browser front, once AOL switches to the Mozilla rendering engine, Netscape and Mozilla users -- and possibly Opera, Galeon and Konq users as well -- will no longer find themselves staring angrily at "Best viewed with Internet Explorer" or "You cannot access all features of this site unless you use Internet Explorer" tag lines -- except, possibly at MSN, which already requires Explorer and Windows Media Player to listen to music. This may be bad for Microsoft, but more Web sites following industry-wide standards is good for everyone else. Maybe the Web Standards Project will finally get some of the respect and cooperation it has deserved all along.
As far as an AOL client for Linux, one Linux-using AOL employee says, "How many Linux people do you know personally who would sign up for AOL if we had a Linux client? I don't know a single one, myself. I have an account with another ISP I use at home with my Linux box, and probably wouldn't use AOL from home even if I could."
The only way AOL could provide a cost-effective Linux client, given its "total support for free" policy, would be to market a real, full-featured personal computer (as opposed to an "Internet appliance") that runs Linux and is preconfigured for AOL. The target market for this computer would not be sophisticated Linux users, but current AOL subscribers who want to replace their current boxes, and it would need to be a very low-cost item to succeed in that market.
Perhaps one of the world's many stalwart Linux entrepreneurs will eventually convince AOL management that an AOL-branded, consumer-priced Linux box is a good idea. Otherwise, AOL will probably stick to the current corporate operating system pattern: Linux in the server room, Windows or Mac on user desktops -- except that AOL-ized desktops will run the AOL browser and its Mozilla rendering engine instead of Microsoft Explorer.
Note: Comments are owned by the poster. We are not responsible for their content.
I've prefered Mozilla to IE since tabbed browsing--especially with a middle-click, load in background in a new tab setting.
Any software changes were made to either automatically eliminate problems, or to make fixing them simple. TS had a good database of problems and solutions. Often customers would call asking for help with problems that had nothing to do with our product...and those calls tend to be some of the longest.
While I understand your frustration, getting the customer off the phone serves one purpose; being able to help the next customer since you are still employed. AOL is big, so small losses won't be missed, but over all the call time must be low. If it isn't, the type of product or service you offer has to change in some way, and it can't always be in dollar cost.
No kidding. I mean, the site works with LYNX for chrissakes, yet they block Mozilla.
Insane.
As for why you can't find this article elsewhere -- in the news business, this is called an "exclusive" or sometimes a "scoop." Robin did his homework, got people inside AOL and Red Hat to talk to him, and reported the story before anyone else did. We like to think that happens pretty often at NewsForge, but in this case, it's a scoop that's a really big deal.
Someone has to be the first to report a piece of news. In this case, it's us.
Grant
When AOL 8.0 is being promoted heavily, I'm sure that there will be plenty of names in the press to link quotes to...though I doubt any of those quotes will have the word 'crap' in them.
"If you use Mozilla, we cannot promise the AOL browser works on future Windows versions , but here's an IE licensing deal that solves such problems...".
You're forgetting that with 1/3 of the US ISP market, AOL has a lot of clout. AOL is no Netscape or Borland, and there would be a HUGE AOL backlash against M$ if they ever tried to sabotage AOL on Windows, otherwise, they would have done it already.
Like most others here, I'd be very happy to see IE's stranglehold broken. AOL using a Gecko-based browser would go a long way towards making that happen.
What I think MS will do, though, is to start spreading FUD about "use MSN, AOL's browser is broken". In other words, they'll start having ads that list the (broken) IE-only sites, and put the spin on it that Joe Average should "get a browser that can render all these sites". It's ugly, but not beyond MS's marketing department. That could backfire on AOL and Mozilla in a nasty way.
""Now think of a Linux client. Either we completely disavow support for it (which is a very un-AOL thing to do), or we try to support every reasonably-up-to-date Linux config in the world. Even with the reasonably-up-to-date caveat, that is a hard thing to do. Where is the market and the demand?" "
Linux is already a small market... it doesn't need further "fragmenting" which is provided by differences in file locations, library versions, etc. People have the nerve to demand Linux client support from AOL, but they are not thinking how difficult it is to support system-level software on it.
Porting AOL to Linux is MUCH MORE than a skin on top of Mozilla! AOL is made for newbies and so provides a lot of system-level handholding. AOL will even do things like prompt you to adjust your system configuration, so you get up and running if your modem is configured. Have any of you folks set up modems on Slackware? If you target the latest libraries in RH 7.2, it won't run in Debian. Look at the platform specific hacks in Webmin, or Linuxconf, etc.
Worst of all, there's no central repository for all this information. Microsoft might have fragmented their Windows platform and there ARE differences, but they are documented for the most part.
The LSB looks like it's gettning closer to being adopted. This should improve things tremendously.
The answer is plain if you just think about it for a second. You can answer it in three letters.
M S N
Microsoft wants a part of the on-line subscription pie (naturally the shape of a Microsoft wedge is a complete circle). Microsoft is a direct competitor to AOL. Would you use as your fundamental technology resources that are fully controlled by your competitor?
Yes, AOL would love it if they could cozy up to a non-competing MS and use the MS desktop to hawk their wares, but they realize that is not going to happen. Their options are to either fight and risk a quick death, or submit and die a certain slow death. I don't know which is preferable from a corporate point of view, but it appears someone at AOL has done the calculus, and the former option looks more appealing.
That is why I am not sure I wanted to see this article just yet, even though I consider
it interesting and very good news. Premature talk may scuttle these kinds
of strategic decisions. Eg. You-Know-Who might step up his FUD or make AOL
management an offer they cannot refuse: "If you use Mozilla, we cannot promise the
AOL browser works on future Windows versions <wink>, but here's an IE licensing deal
that solves such problems...".
Keep in mind that the change from IE to Mozilla is not the same as hiring RedHat to support Linux servers. There may be some overlap, but if there is any I can't see it.
As for the Mozilla support, AOL/TW owns Netscape so they have ample support for the switch.
RedHat will probably get the contract, but even if they don't there are other companies with Linux experts that would be very willing to step in.
I doubt it. When Windows desktop real estate was up for offer, this was possible. Those talks broke down over a year ago and I've seen no movement on either end...even when AOL/TW switched Compuserve customers over to Mozilla.
Microsoft is threatend by AOL/TW. AOL knows MS will do just about anything to run the AOL part of AOL/TW out of business.
My only surprise is how conservative AOL has been in making the switch. True a change like this should be at a major revision, like 8.0, but I'm surprised that there haven't been rumors of live beta code available for testing.
A reasonable time to float a beta release is a few weeks after the Mozilla 1.0-based Netscape 6.x appears to see if there are any major snags. Rumors of AOL's 8.0 software will probably start appearing 2-3 months from now; May/June.
Re: "Even so, will we all not be better for it all."
I'm guessing you ment something like 'Either way, we will be better off because of AOL's considering/threatening a switch from IE to Mozilla.'
As for Linux support, why not offer a "Beta trial" to existing AOL customers. This can be kept low key. It can guage the interest in Linux and ultimately produce a quality Linux AOL client. I am not sure why they are reluctant to add Linux. It is up and coming technology and is already quite solid. They have the Linux Mozilla code and the rest should not be all that difficult.
My little sister uses Linux, she doesn't know jack, and is not used to configuring her computer. I set up her machine, I send her update disks, I login via SSH and make any changes that confuse her.
When she originally got her machine, I sent her a video tape showing her exactly how to use her machine...including mouse movements and important tips. She still calls me up and asks how to do even basic tasks.
That would certainly solve the cross platform problems, and the current JDK would work fine for such an app. It would really show off client-side Java, which still has it's share of sceptics.
I guess that'd make too much sense...
Java AOL client...
Posted by: Anonymous Coward on March 11, 2002 11:31 PMThat would certainly solve the cross platform problems, and the current JDK would work fine for such an app. It would really show off client-side Java, which still has it's share of sceptics.
I guess that'd make too much sense...
#