The blog of a Network Analyst who plays around with many things open source when he is not feeding his MMORPG addiction.
RSS icon Email icon Home icon
  • Games and Vibes

    Posted on February 29th, 2004 Bruce 4 comments

    Bleh. I’m exhausted and going to bed. Started the day by hosing my desktop system by installing Fedora Core 2 rawhide (development branch). I managed to get most of it working but I’ll need to figure out how to install an mp3 plugin for rawhide’s version of gstreamer and get some minor blog tools installed. Projects for tomorrow.

    After spending a couple hours on the desktop system I then spent the rest of the day with my Linkshell getting level 34 Bard and getting Rank 4. The mission took us about three hours to do (instead of the 1.5 hours the walkthrough suggests). Add on to that the time I spent trying to catch up on my missions in a hurry so I could do it with everyone else and that was one long day spent in front of a computer.

    My ass is giving me those “get the hell off me” vibes. Yep.

  • OJ Isn’t The Only Celeb To Wear Ski Masks

    Posted on February 27th, 2004 Bruce 2 comments

    Yahoo! News / AP Article: “Michael Jackson was pulled over by police and asked to identify himself after shopping at a Wal-Mart while wearing a ski mask.”

    My response: “Bahahahaha!”

  • Do Babies “Plump” When You Cook Them?

    Posted on February 27th, 2004 Bruce 1 comment

    Hmm… I have to congratulate CNN.com on this annoying little article. Of course the best part was they actually quoted the babies idiot father.

  • Fable Article

    Posted on February 27th, 2004 Bruce 1 comment

    You may wish to check out a very interesting article on the upcoming xbox game Fable. I guess we’ll find out if it’s as good as the article says or if it’s pure bullshit this summer.

  • Server Upgraded

    Posted on February 23rd, 2004 Bruce 1 comment

    I decided to chip in a few extra dollars a month to upgrade the virtual server from 64mb to 128mb of memory. I guess 64mb just isn’t enough to run a web server using python cgi scripts, a mail server, virus and spam filtering, etc.

    So, in summary, the comments pages shouldn’t be slow as hell anymore. Weeeee….

  • More New Paltz Fun

    Posted on February 23rd, 2004 Bruce 2 comments

    Live Journal has a “newpaltzjunkies” interest group made up of SUNY New Paltz college students. It even has a handy RSS feed.

    Right now there are a total of 205 Live Journal users which list “New Paltz” in the Interests part of their profiles.

    Feedster lets you search a large library of RSS feeds and generate an RSS feed based on feed entries that match. Look at all the fun at http://www.feedster.com/search.php?hl=en&ie=UTF-8&q=new+paltz&btnG=Search&sort=date.

  • How New Paltz Are You?

    Posted on February 23rd, 2004 Bruce No comments

    A little quiz I ran into: How New Paltz Are You?

  • Quiznos Commercials

    Posted on February 23rd, 2004 Bruce 1 comment

    The recent Quizno commercials apparently are the work of the demented person who runs http://www.rathergood.com/. I have to congratulate this person for a commercial that is so annoying I usually change the channel as soon as it comes on.

  • Root Beer

    Posted on February 23rd, 2004 Bruce No comments

    Mmmm… Root Beer.

    For some reason I had a craving for Root Beer this weekend. This is probably the first time I’ve had it in years. Mmmmm.

  • A discussion about Red Hat

    Posted on February 22nd, 2004 Bruce No comments

    The following is a lengthy reponse to a discussion on a LUG mailing list about Red Hat’s handling of the transition from Red Hat 7.x/8/9 to Fedora. Since I took the time to write it I might as well post it up on here. (I didn’t want to spend too much time on this so there are probably grammar and spelling mistakes)

    The “only bad thing” that Red Hat did in my opinion was not to declare a support policy ahead of time with “set in stone” deprecation dates and tried to put too much on its plate. They didn’t say “If you buy this it comes with 24 months warranty”, they always seemed to keep saying “as long as it seems feasible to support it” or other non-answer. They found themselves being asked to maintain over 5 revisions of the “base” Red Hat operating system over the course of 2+ years by a “customer” base of which only a small percentage gets around to buying a token copy.

    I’d really like to see the train of thought that says Red Hat should have continued supporting a product line for multi-year periods when the profit model was shown to be a failure by every other commercial Linux vendor.

    I’m sure we all know the efforts of Mandrake to recover from bankruptcy, of which was a shift away from the base distribution onto ‘renewable’ sources of income such as MandrakeClub and its fledgling Enterprise versions. SuSE, which has been in the red for a while and required injections from IBM, The German Government and other sources had been trying to emphasize its SLES version before it recent acquisition by Novell.

    Before the announcement of Red Hat 9.0 and its first official policy on product lifetimes for its base distribution (something that it had never set in stone before) Red Hat must have realized it had two product lines serving two separate customer groups: The Enthusiasts and its Business, Education, and Government markets.

    The Enthusiast market demands a new version every six months or so. They want relatively cutting edge versions and they want it for cheap. They, are for the most part, are content to upgrade every six months or every year. Those machines that do not get upgraded tend to linger around and stay unpatched and unmaintained. Basically a thorn in the side of your average .edu administrator.

    The Business, Education, and Government markets had major problems with an operating system that has a six month release cycle. They are deploying Linux to run a mix of “open source” and proprietary packages. The ISVs and proprietary software developers they depended on to provide some of this software hated Red Hat’s release cycle. What they ended up doing was picking an arbitrary release to support and told customers it was their problem if they didn’t use that arbitrary version.

    This causes pain to the customer who is told that the brand new piece of proprietary software he needs requires another piece of proprietary software that requires Red Hat X (where X is any one of the 6 versions produced in the last three years).

    For instance, Blackboard, a piece of software that is used by many .edu’s, required Oracle 8i up until a month or two ago (of course Blackboard upgrades must be scheduled and are often painful in themselves). Oracle 8i was originally targeted for Red Hat 6.2. Blackboard (until its .edu customers demanded and _dragged_ it into supporting RHEL) preferred Red Hat 7.2. Installing Oracle 8i on Red Hat 7.2 is a pain in the ass that involves compatibility packages, vendor patches, install script editing, fighting with Java JVMs, etc. Blackboard has its own issues but you get the idea. This pain gets played out quite a bit in many different ways.

    Even without introducing the world of proprietary software and third party support contracts into the mix, even upgrading production systems more then once every two or three years gets painful fast. I can think of quite a few upgrades that required major reworking of configuration files, testing of new configurations, the exporting and reimporting of databases, etc due to incompatible formats, etc.

    So as Red Hat tries to push into the business, education and government markets it keeps hearing over and over again that they, and the ISVs they depend upon, want longer supported life cycles with various degrees of support contracts, hardware certifications, etc. Most of all they want the pain of having to deal with all these different versions to stop. Three years may seem like a long time to Linux Enthusiasts but large organizations do not share this view and do not have the man power and resources to keep upgrading everything because a new version is out. (Notice how many Windows 98 machines are still around? For a very good reason.)

    Red Hat obviously looked at its six month development cycle for its base version, and its two customer bases and realized the two customer bases demands were at odds with each other, and of course there was an opportunity to profit off this by launching a separate product line with a renewable yearly source of income. If you have not noticed, this trend is already taken place industry wide. All of the major pieces of software used by organizations already follows this licensing model. As much crap as Microsoft got for their new license programs, they were not the first. Even Mandrake and SuSE are getting into the game in their own ways.

    So Red Hat now has its Enterprise line on its plate to deal with. These customers were guaranteed a five year support cycle for each release. Red Hat now has to deal with the fact it has two product lines, of which most people don’t really understand.

    Red Hat now has to deal with multiple releases on multiple processor families with kernels, glibc, etc modifications that are different between each other. On top of this Red Hat is expected to continue support for its Enthusiast versions which bring in a token amount of money compared to its Enterprise line.

    Red Hat’s engineering team was left trying to support a crapload of slightly incompatible permutations of Linux. They try to ease the pain of the task ahead of them by trying to group base line Red Hat releases into families based on kernel and glibc. Eventually I’d guess some people within Red Hat sat down and had a meeting.

    I can only guess what was discussed at the meeting. My best guess would be it was along the lines of “This is crazy… we can’t keep doing this. Supporting the base line Red Hat version isn’t worth the profit we are taking in for it. We are chewing engineer time that would be better spent on the many projects we have internally and we support externally.” Of course if some people are to believed the meeting involved the sacrificing of some goats and customers. Feel free to insert your own theory. :)

    Red Hat announces the “shortened” life time (aka they finally set stable “set in stone” death dates) and deals with the uproar. Quietly it works with its larger customers (you know, the ones that actually bought the boxes, not just burned the ISOs then bitched about it on Slashdot) to explain its decision and encourage them to jump on the Enterprise train. Obviously not everyone agrees with this, but that is their choice, the nice thing about having competition in the field.

    Other customers (such as many .edus) find themselves with entitlements for RHEL products for the remainder of their up2date periods for the registered box sets of Red Hat 7.x and 8 they had bothered to register with RHN. (Notice the pattern: Paying customers that Red Hat knows about are given a bridge to its Enterprise product line… insert your own conspiracy theory.)

    While this is all happening Red Hat announces the Fedora project. This project has the following benefits for Red Hat and others:

    • It is viewed as more of a community product then the Enterprise product. This makes it easier for customers to decide between which product they want. Of course not all of the former box set customers are too happy with Red Hat and switch to another vendor. Another great thing about having competitors offering similar products.
    • The distributions new name is “Fedora”. Red Hat can now strip out all of its Trademarks and Logos. The new distribution is now distributable and sellable by third parties using the Fedora name and not trademarks that Red Hat has to protect. Remember all the commotion when Red Hat went after CDR sellers and made them stop calling the CDRs “Red Hat Linux”? Red Hat promised a long term solution for it and this was it.
    • The target market for this product is made crystal clear: It is for Enthusiast. It will offer the cutting edge in technologies that Red Hat is working on and frequent releases that makes the cutting edge kiddies all hot and bothered over. ;)
    • The support cycle for Fedora and RHEL is crystal clear. Red Hat even indirectly encourages the formation of Fedora Legacy (http://www.fedoralegacy.org) through the Fedora.US camp (which is working to merge with Red Hat’s efforts) for the errata support of older releases past its defined support life-time. Basically, the releases are “supported” with bug fixes and security fixes as long as someone decides it is worth their time and resources to do it. If no one decides this, then the product “dies”. Welcome to the Open Source world.
    • Red Hat uses Fedora as their development branch for Red Hat Enterprise Linux. Red Hat continues to devote significant engineer resources both to the inclusion and maintenance of upstream software and software that it writes itself and contributes to the open source world.
    • Red Hat attempts to throw an olive branch to the world by changing its formerly “we release it and you have no say” product line into a community project. Unfortunately this is proving to be a slow process but it is happening. (I can elaborate on this more if people want but this is getting way too long already…)

    So what does Red Hat get out of this? It gets a product line that is easily recognizable outside of “The Community” and isn’t clouded by the confusion that was created by having two disjoint product lines.

    What does the community get out of this? A fully 100% open source free from the trademark problems that plagued earlier releases. It gets a level of access to Red Hat developers not previously available through the very active Fedora lists, a voice in the evolution of this new distribution (more on this in a second) and a test bed for the latest technologies with a defined and frequent release strategy. [Linus and the SE Linux .mil people have made many comments on the fact they view the Fedora 2 release as the first real tests of Linux 2.6 and SE Linux. Almost to the point that there was some baseless trolling on the SE Linux mailing lists about the project becoming Fedora focused.]

    Now chances are many of you are bitter about Red Hat’s choices or simply decided that Red Hat and Fedora are not for them anymore. The latter are welcome to that opinion and I like the fact if I decide I no longer like Fedora or Red Hat that there are alternatives in the wings.

    To those people who were bitter the following will not be a popular but more people need to understand it:

    In the Open Source world if you do not contribute money or code (or be able to convince someone influential enough to contribute money or code) your voice is worthless. All the pissing and moaning in the world will not get you what you want unless you are willing to help do it yourself or pay to have it done.

    Open source is a meritocracy not a democracy.

    To the tiny minority that did pay for the boxed set for each and every machine they installed it on, then you have every right to feel pissed and burnt if you were counting on that product line. Your decisions are perfectly understandable.

    Unfortunately the vast majority of people I’ve met (both online and real life… hell I used to be part of this camp) who were bitter about the arrangement were people who downloaded ISOs and used those or bought a single copy of the boxed set for the media and then went around and installed it on 10 machines and gave copies of the cds away. It is not wrong or unethical to do that. Just don’t be surprised when something that you are getting for free or near free goes away. Just because open source is “free software” doesn’t mean it doesn’t take man power (and hence money) to maintain it.

    My motto for Open Source is “if you are demanding something and you not willing to contribute enough money, code, or time to help it get done, please feel free to get bent.” You don’t have a voice if you don’t contribute. Period.

    Just because there are alternatives out there doesn’t give you the right to piss and moan. Don’t be surprised someday when some volunteer contributer to Debian or Gentoo bite your head off when they think you are a demanding whiner who doesn’t contribute anything.

    Yes, I am a troll. ;)