You all probably know that the most important thing on any multi user system is a pretty MOTD. Between some other things in the past couple of weeks I decided to refresh the MOTDs for the galaxy and Seawulf clusters. I discovered 2 awesome applications while compiling the MOTD.
First is a jp2a, it takes a JPG and converts it to ASCII and it even supports color. I used this to render the milky way as part of the galaxy MOTD. While this tool is handy it needs some assistance, you should clean up and simplify the JPGs before you try to convert them.
The second tool is a must for any form of ASCII-art editing. Its called aewan (ace editor without a name). It makes editing a lot easier, it supports coloring, multiple layers, cut/paste/move, and more. Unfortunately it uses a weird format and does not have an import feature, so its PITA to import an already existing ASCII snippet -- cut and paste does work but it looses some information -- like color.
Aewan comes with a sister tool called aecat which 'cats' the native aewan format into either text (ANSI ASCII) or HTML. Below is some of my handy work. Because getting browsers to render text is PITA I decided to post the art-work as an image.
I also wrote a short cronjob which changes the MOTD every 5 min to reflect how many nodes are queued/free/down
One more resource I forgot to mention is the ascii generator. You give it a text string and it returns in a fancy looking logo.
Finally when making any MOTDs try to stick to the max width of 80 and heigh of 24. This way your art work won't be chopped even on ridiculously small terminals.Debian LILUG News Software 2008-03-02 23:41:22
I haven't posted many clustering articles here but I've been doing a lot of work on them recently, building a cluster for SC07 Cluster Challenge as well as rebuilding 2 clusters (Seawulf & Galaxy) from the ground up at Stony Brook University. I'll try to post some more info about this experience as time goes on.
We have about 235 nodes in Seawulf and 150 in Galaxy. To boot all the nodes we use PXE (netboot), this allows for great flexibility and ease of administration -- really its the only sane way to bootstrap a cluster. Our bootstrapping system used to have a configuration where the machine would do a plain PXE boot and then, using a linuxrc script the kernel would download a compressed system image over TFTP, decompress it to a ram-disk and do a pivot root. This system works quite well but it does have some deficiencies. It relies on many custom scripts to maintain the boot images in working order, and many of these scripts are quite sloppily written so that if anything doesn't work as expected you have to spend some time try to coax it back up. Anything but the most trivial system upgrade requires a reboot of the whole cluster (which purges the job queue and annoys users). On almost every upgrade something would go wrong and I'd have to spend a long day to figure it out. Finally, using this configuration you always have to be conscious to not install anything that would bloat the system image -- after all its all kept in ram, larger image means more waste of ram.
During a recent migration from a mixed 32/64bit cluster to a pure 64bit system. I decided to re-architect the whole configuration to use NFS-root instead of linuxrc/pivot-root. I had experience with this style of configuration from a machine we built for the SC07 cluster challenge, how-ever it was a small cluster (13 nodes, 100cores) so I was worried if NFS-root would be feasible in a cluster 20 times larger. After some pondering over the topic I decided to go for it. I figured that linux does a good job of caching disk IO in ram so any applications which are used regularly on each node would be cached on nodes themselves (and also on the NFS server), furthermore if the NFS server got overloaded some other techniques could be applied to reduce the load (staggered boot, NFS tuning, server distribution, local caching for Network File systems). And so I put together the whole system on a test cluster installed the most important software mpi, PBS(torque+Maui+gold), all the bizarre configurations.
Finally, one particularly interesting day this whole configuration got put to the test. I installed the server machines migrated over all my configurations and scripts halted all nodes. Started everything back up -- while monitoring the stress the NFS-root server was enduring, as 235 nodes started to ask it for 100s of files each. The NFS-root server behaved quite well using only 8 NFS-server threads the system never went over 75% CPU utilization. Although the cluster took a little longer to boot. I assume with just 8 NFS threads most of the time the nodes were just standing in line waiting for their files to get served. Starting more NFS threads (64-128) should alleviate this issue but it might put more stress on the NFS-server and since the same machine does a lot of other things I'm not sure its a good idea. Really a non-issue since the cluster rarely gets rebooted, especially now that most of the system can be upgraded live without a reboot.
There are a couple of things to consider if you want to NFS-root a whole cluster. You most likely want to export your NFS share as read-only to all machines but one. You don't want all machines hammering each others files. This does require some trickery. You have to address the following paths:
You cannot mount this to a local partition as most package management systems will make changes to /var and you'll have to go far out of your way to keep them in sync. We utilize a init script which takes /varImage and copies it to a tmpfs /var (ram file system) on boot.
This is a pain in the ass I don't know who's great idea was to have this file. It maintains a list of all currently mounted file systems (information is not unlike to that of /proc/mounts). In fact the mount man page says that "It is possible to replace /etc/mtab by a symbolic link to /proc/mounts, and especially when you have very large numbers of mounts things will be much faster with that symlink, but some information is lost that way, and in particular working with the loop device will be less convenient, and using the 'user' option will fail." And it is exactly what we do. NOTE autofs does not support the symlink hack, I have a filed bug in the debian.
- /etc/network/run (this might be a debianism)
We use a tmpfs for this also
We mount this to a local disk partition
All in all the NFS-root system works quite well I bet that with some tweaking and slightly more powerful NFS-root server (we're using dual socket 3.4Ghz Xeon 2MB cache and 2GB of ram) the NFS-root way of boot strapping a cluster can be pushed to serve over 1000 nodes. More than that would probably require some distribution of the servers. By changing the exports on the NFS server any one node can become read-write node and software can be installed/upgraded on it like any regular machine, changes will propagate to all other nodes (minus daemon restarts). Later the node can again be changed to read-only -- all without a reboot.
Debian LILUG News Software Super Computers 2008-03-02 13:25:11
MythTV Lives On
Many of the LILUGgers will remember that I talked about uncertain times coming up in the future of MythTV. One of the companies (Zap2it) that provided all the channel information was pulling the plug on the service. Well I just heard great news.
A bit over a month ago I sent an email to Zap2it thanking them for their great service and support of the MythTV community.
From: dotCOMmie [mailto:####@#####.###]
Sent: Monday, June 25, 2007 9:44 PM
Subject: Thanks for the great service.
I'm relatively new to the mythTV community, and it is how I got to know
Non-the less I think you were a crucial part to the tremendous growth of
this community, and I'm saddened by your recent decision to stop
providing the channel listing service.
The reason I'm emailing you is to convey my deepest thanks to your
company for providing such an excellent service to the hobbyist
community. I often wish we had more companies like you.
And here is the reply I received today:
From: "Roberge, Andy" <########@#######.###>
To: "dotCOMmie" <####@#######.####>
Date: Tue, 7 Aug 2007 21:17:44 -0500
Subject: RE: Thanks for the great service.
August 8, 2007
Zap2it Labs received many emails inquiring if television listings could
be provided on a paid for basis once the current service is
discontinued. Today we are pleased to announce an agreement that will
allow for many of you to continue to have access to your personal
television listings data.
In collaboration with Schedules Direct, a non-profit organization
created by founding members of MythTV and XMLTV, an agreement has been
reached that will continue to support the open source and "freeware"
communities. As of September 1, 2007, there WILL BE an alternative
television listings source for certain Zap2it Labs users who become
members of Schedules Direct, which includes a membership fee.
While Schedules Direct will continue to support the open source
community and the users of "freeware", it will not support users of
"commercial" products, such as DVRs, that were purchased from either a
retail outlet or a company that used Zap2it Labs as its television
For those of you who lived by the original spirit and intent of Zap2it
Labs and wish to continue to have access to listings, we encourage you
to visit Schedules Direct at http://www.schedulesdirect.org to set up
your personal Schedules Direct membership today!
I honestly have to say I wasn't expecting a reply let alone such good news. Thank you once again, Zap2it.Lilug MythTV News Software 2007-08-07 22:37:12
LIRC is a software package under linux which allows you interface with remote control/controlled devices. LIRC is pretty much a must for any 1/2 decent MythTV configuration.
For my Myth setup I use LIRC both to change the channels on the cable-set-top-box and as a way to control the mythtv interface from the couch. Although this is a quite common configuration its annoying to get working.
The first thing you have to decide when setting up LIRC is what hardware you want to use. You can build your own receivers/transmitters but the simple plans make for quite crappy and unreliable devices; for something more sophisticated the cost of parts adds up to exceed the cost of kits/ready-to-use devices.
I had a (X10 based) RF serial receiver and remote (that I got a while back with my Nvidia PC cinema card). It worked with better LIRC than it ever did under windows. To control the set-top-box I first got an iguanaworks USB transceiver but it would not work since it only transmits at 36khz (It can be flashed to transmit at 58khz with a non-existing utility) and all the devices I needed to control only worked at 58khz. Money down the drain. So I decided to try again, this time I got the Serial Iguanaworks transceiver this one interfaces with LIRC more like the home-made transceivers except it has greater range (thanks to a .3f capacitor (think battery) which stores energy for transmissions).
Alright so I'm thinking I have the hardware configuring should be a breeze. I already had the controlling software installed, all I needed was to compile the drivers. I downloaded the debian driver source package it looked all very nice and neat, it allowed me to select the drivers I want and even attempted to compile the drivers automagically.. except it failed. The sources it provides are too old and were no longer compatible with my kernel. No big deal, I'll compile the vanilla drivers from LIRC -- wrong.
LIRC can't be compiled with just any combination of drivers you want, the configuration scripts compile either any ONE driver or all of them. No big deal, I thought, I'll compile all and install only the onces I need.. except all the drivers don't compile. Compilation broke on some driver that I didn't need. So I decided to hack the config scripts a bit. I downloaded the CVS version of LIRC opened the configure.in file and around line 1207
if test "$lirc_driver" = "all"; then
Trimmed down the list of drivers to only the ones that I needed. I then ran autoconf to generate all the needed Makefiles and ran ./configure --with-driver=all --with-port=0x3f8 --with-irq=4 --with-timer=65536 --with-x --with-transmitter && make && make install and things built correctly with only the drivers I wanted.
From then on configuring LIRC was a breeze, I modified the debian /etc/init.d/lirc script to use start 2 lirc daemons, one for each driver and configured them to talk to each other.
Finally I made my lircd.conf and lircmd.conf using irecord and configured MythTV, xorg and channel changing script. YAY, working mythbox.
Brief overview of all the programs and devices that make up my mythbox
A/V Hardware: Nvidia MX440 (vga/svideo out), Happauge150 (rca audio/svideo in), CHAINTECH AV-710 (optical audio out), RCA dvd/audio system
Remote controlled devices: RCA TV, Scientific Atlanta Explorer 4200 (cable box), Nvidia branded X10 RF remote
The last problem I had was the cable box being off while mythtv was trying to record, Its a nasty one. But it turns out the cable box has this nice feature where it will turn on when any numerical key is pressed on the remote (can be enabled in the settings menu). So when mythtv changes channels the cable box is either already on or is turned on auto-magically.
More of my config files.Debian LILUG MythTV Software 2007-06-29 00:04:25
Chimei 22" Nvidia
I thought the days of modelines in xorg (and linux in general) were over but I guess I'm wrong. The last 2 monitors I configured I had really difficult time with. One needed needed just a modeline but the other needed nasty config hacks. The first configuration was a Dell 21" monitor with a i945 graphics card and the other a 22" Chimei CMV-221D/A with an nvida GeForce FX 5200 card.
The Chimei monitor autodected just fine over VGA but was fuzzy and wavy, and hooking it up over DVI, the nvidia card did not want to drive it over 800x600 (instead of native 1680x1050). So I had to get down and dirty with the X configs.
Anyway here are the appropriate sections from my xorg.conf file for the Chimei (I'll post the Dell ones later)
Identifier "Generic Monitor"
Identifier "nVidia Corporation NV34 [GeForce FX 5200]"
Option "NoLogo" "true"
#NOTE this is probably dangerous only use this line with appropriate Modeline
Option "UseEdidFreqs" "false"
Option "ModeValidation" "NoMaxPClkCheck,AllowNon60HzDFPModes,NoVesaModes,NoXServerModes,NoPredefinedModes"
Identifier "Default Screen"
Device "nVidia Corporation NV34 [GeForce FX 5200]"
Monitor "Generic Monitor"
Modes "1680x1050" "1024x768" "800x600" "640x480"
Modes "1680x1050" "1024x768" "800x600" "640x480"
Modeline "1680x1050 (GTF)" 154.20 1680 1712 2296 2328 1050 1071 1081 1103
EnjoyDebian LILUG Software 2007-06-27 23:18:46
Browsers -- I hate them
I hate browsers every single one that I've used. Every browser out there is a pathetic failure when it comes to user interface. Right now my favourite browser is iceweasel/firefox but in my book it doesn't have much going for it.
The browsers have a love for pop-up-dialogues. It's getting a little better but not good enough. I remember when in firefox if you mistyped a URL it would pop a dialog box "Server not found." So you'd have to take your hands of the keyboard and hit OK and then put the cursor back to the address bar and try again. Why does the browser need to confirm with me that I mistyped something? Now this is no longer a problem; when you go to non-existing page you'll get a message insider your browser pane saying that server cannot be found. This is great but I believe that NOTHING should pop-up without the users intent
Say, for example to search for something on google and you get a link to a mailing list. I've seen a few mail-list archives where they use self-sign signatures (https) so you get a pop-up dialogue saying that the page is not kosher. WHY?! Its not a page I care about for security; in fact most pages I visit I don't care much if the anyone spies in on what I read. I think this warning should be brought up where it can be ignored without any user interaction. For example a drop down bar with a message (like those pop-up blocked notice). Heck you can even turn the whole browser panels and things RED so even the most senile users will notice something strange is up. And maybe the first time the user comes across this error it should pop a dialogue explaining why the browser miraculously turned red.
Users hate dialogues if it has more than 200 or so characters in the message a majority of the users won't even read it, they will in a robotic-type fashion click on some button until the dialogue will disappear. So, just stop with the pop-up dialogue boxes they are annoying and not useful. If your program needs to constantly pop things up for user to select then you have failed user interface design.
Iceweasel/Firefox has this awesome feature where you can scale the page fonts. Its incredibly handy when you come across a web 2.0 website with 2 point font (fucking web designers, readability first style second!! STOP IT!!). Now this is all fine but I am tired of always manually adjusting the fonts per website. Fortunately there is other great feature (Edit>preferences>content>font&color>advanced>minimum font size) where you can set the minimum font size. Well you'd think this is the best thing since sliced bread (figure of speech, I hate sliced bread too but thats for another day) but there is a tremendous flaw with this feature. When you select a minimum for of size, say, 8 every font thats less than size 8 will be turned to 8 all larger fonts will not be affected. This sounds great in theory but horrible in practice, if you got to some heavily stylized graphics your setting will send a lot of fonts out of boundaries. So you'll get overflowing menus, notices and all that other jazz. Its so annoying its that its not usable. What the browser should do instead is scale all the fonts on the page. Say the smallest font on the page is of size 5 then 8-5=3 so increase EVERY font on the page by 3 points, kind of like what happens when you use manually adjust font size (view>text size>increase).
Stupid menu-bars. Every browser is full of them. You have the status bar on the bottom the menu bar, search bar, tab bar and bookmark bar on the top, WTF?! When I use the browser I want to see the webpage not the static content of the browser. STOP stealing my real estate. So I suggest you disable the bookmark and the status bar. And you'll scream BUT I want the functionality of my status bar; "I want to know where the link points that I am about to visit." Well so do I, I hate the bar but like the functionality it provies, but there is nothing to say that the functionality can't be moved. Say when you move your mouse over a link your address bar displays the address of the link, and as soon as you move away from the link the address bar goes back to displaying the address. As for the load status, I've found this great plugin called fision which takes from a safari feature, shows the progress of the loading in the background of the status bar.
The great menu bar, its immune from any customization. I just sits there, does nothing most of the time, face it how often do you use it? While its very useful its not needed all that often (maybe once a week) so why is there not a feature where it can collapse into a expandable menu (kind of like the start button on windows or kmenu in kde) And when you click this monster it would just appear. Now allow this menu button to be place into any other panel and forget about. What a real estate saver.
Who Wrote This Shit
Portmap by default listens to all IP addresses. However, if you are not providing network RPC services to remote clients (you are if you are setting up a NFS or NIS server) you can safely bind it to the loopback IP address (127.0.0.1)
<Yes> OR <No>
Maybe I'm slow or something but I really hate this prompt in debian. Which is accompanied by the installation of portmap. Seems like you need a degree in english logic to figure out what you need to select. If you run NFS and NIS and are Confused the hell out by this prompt just select NO.
UPDATE: Just because you select NO doesn't mean that debian will actually not bind RPC to portmap. You might want to run dpkg-reconfigure portmap again and make sure it did the right thing.. I got a nasty surprise the day after .. when 2 of the NFS servers stopped mounting. Filed bug reportDebian LILUG Software WWTS 2007-05-25 21:21:52
qmail, a love - hate relationship
After years of procratination I finally got around to whipping my mail system back into shape. Its quite common qmail+vpopmail+courier-imap+courier setup. What I really wanted was spamassassin and some more features in the web based front end (ilohamail).
I've tried to setup spamassassin a couple time already but each time as I started reading howtos all the vigor would dwindle away. The whole qmail configuration and philosophy is great, take all the pieces you need arrange them together (with some help from duct tape and wd-40) -- and voila a full-fledged mail server! Except one day you'll pass a point where you can barely keep track of all the pieces that keep your mail together.
So as I was reading how to get spamassassin integrated into qmail I thought, "Qmail sucks I hate qmail.. its way too complicated." A little later (think HOURS) I had broken the delicate chain of apps and inserted qmail-scanner and spammassassin (spamd and spac) into the mix. Amazing things still worked and now my email had special X-Spam-Status tag (YAY!).
Qmail and me have a true love-hate relationship. When it works and you don't need anything extra from it, then, its great. But as soon as you get this nagging feeling "I wish I had feature X" you start slowly hating qmail -- untill you fix it up and make it all pretty again.
As a sidenote -- pay no attention to the fact that ilohamail 9.X is a beta version. It works great and the new features are way worth the migration from 8.X. Email filters, imap over SSL, gpg, public calendar... the list goes on and on.
So once again -- with the help of bind9 (SPF), spamassasin, qmail scanner, ilohamail 9.X -- my mail system is modern!LILUG Software 2007-05-15 23:08:31
I guess I in some kind of "documentation" mood. I've written up 2 linux how-tos one for using PPTP under linux and the other is an article for tuxmobile.org about installing Debian on a Dell XPS M1210.
LILUG News 2006-09-20 20:40:26
I've spent the whole summer working at BNL developing a ethernet based FPGA programmer called n3c. Its based on a coldfire m68k chip an awesome platform. As soon as I get an ok to post (part of the) the sources I will. Now I have a few things in progress a cool PIC18F controlled fountain, some linux how-tos and school.2006-09-16 19:47:58
Open Source Today (WRT102 research paper)
In the mid 1980s a new movement came about, a movement to free computer users from the restrictive model of software development. Up until this time most software developers (especially those in corporate environments) were intent on keeping their software source code in the dark; this way no one was able to modify or "steal their software." While today this idea is still the dominant one, the tide is changing. The concept of Open Source is becoming more popular and accepted with each day. One fundamental question remains: Is Open Source Software (OSS) really a god send, will people and companies accept OSS solutions over Proprietary ones?
Sami describes Open Source as concept of developing and distributing software which is the direct opposite of the more common and familiar Proprietary Software (2). More often than not, when a user receives a (proprietary) computer program they get a compiled version (also known as the binary or an executable), it contains a series of computer instructions which are generally uninterpretable by humans. This binary is not what the the developers use to write the program; it is instead a translation from a computer-language (source code) to pure machine-instructions. A common analogy of executables is a "black box" which, when given certain input, will return an output. The source code however is a blueprint describing what is contained in this black box. Unlike proprietary software, when a user receives Open Source Software they also receive a copy of the source code. With this source code the user can theoretically modify and customize a program to their liking.
There are two goals which the Open Source movement is trying to address, both of which bring the software development closer to the user. From a practical approach, when compared to proprietary software, OSS evolves at a phenomenally fast rate. Since many people can analyze a programs inner structure for bugs and security flaws, patches and updates are released faster. It is for this reason alone that many people choose OSS solutions instead of the proprietary ones. From an ethical perspective, some Open Source developers believe that they have a responsibility to society in providing better software. Unfortunately sometimes this responsibility is hindered by legal means, it is not uncommon for a company or developer to file for a patent on algorithm. In essence this prevents or limits other developers from implementing this or similar algorithms in their programs.
Lately the concept of Open Source has turned into a political border-lining religious philosophy. There are two main churches (or cults) of Open Source Software, Open Source Initiative (OSI) and the Free (Libre) Software Foundation (FSF). While the underlying idea of these two organizations are similar -- the source code should always be provided with the binaries -- they are not homogeneous. Richard Stallman of the FSF describes the difference the best, he claims that FSF and the OSI differ in their values. Unlike the FSF, OSI approaches this issue from a practical perspective rather than an ethical one (55).
Today the user base of Open Source Software is still small fraction of all computer users, but it should not be ignored. Over the past few years the user base has been growing dramatically. If this kind of growth rate is sustained Open Source users will soon become a significantly sized group. However, one should not jump to the conclusion that while many people are jumping onto the OSS bandwagon that they are also welcoming the the aforementioned religion. Although it is virtually impossible to measure; many Open Source users are not activists or converts they simply use Open Source solutions because they are either more convenient or, perhaps more importantly, because they often gratis.
Perhaps you have heard from a friend how Open Source solutions are better and so much more secure, or maybe you have recently seen an IBM commercial bragging how they are now providing secure Linux based solutions. If OSS is so much better why isn't everyone using it? The simple truth is that apart from a few projects (ie: Firefox, OpenOffice.org) Open Source is not very user friendly and generally requires extensive knowledge of computer/software systems. As far as security goes, Open Source alternatives are not per se more secure. There are many reasons why Open Source Software programs fare better when it comes to viruses and intrusions: they are not as an attractive target because there are generally fewer of them; they are patched more quickly when security flaws are discovered (ie: ping of death exploit); and finally, as pointed out by Ferris, people running advanced Open Source systems (ie: Linux) are generally more aware of security and learn to avoid compromising their machines(40). But of course security on Open Source solutions is not all hype. Many alternatives to proprietary solutions such as Apache and Firefox have been time and time again been proved to be more secure.
The most promising area for OSS today is in the corporate environment, as companies require ever more dynamic solutions to their problems proprietary software can only go so far. Generally it is very hard to integrate several proprietary systems together into a single solution, and the fact that you are often not able to customize proprietary software which you bought to suit your task doesn't help you much. For an systems administrator it is generally more important to do exactly this. For example if you want to use one package for a mail traffic handler and a totally different system for mail box handling you could, theoretically, achieve just this. If you wanted to do this with proprietary software you'd probably be out of luck, its highly unlikely that company A which makes a mail system would like to make their mail handler compatible with the mailbox handler of company B.
As mentioned earlier there is a good number of companies that embrace Open Source Software, but there is also the other side. Recently several high profile companies (Microsoft, SCO) have gone on tremendous slandering campaigns to try to kill the the OSS movement while it is still developing stage. A question emerges: if some are able to turn a profit while producing "free" software why cant others adjust their business plan to the new wave? This is generally not a question of economics alone but also, one of politics.
Companies which provide Open Source software are usually more interested in selling solutions then programs. IBM for example sells specialty hardware with integrated OSS packages which together aim to make it easier for administrators to implement the solution into an existing system. As a company, IBM is not concerned about selling software. It knows that the software alone, without the special hardware, is not of interest to the public. Other OSS oriented companies (ie: RedHat, Novell) make their money by either selling support to a product which they "give away" or alternatively they provide a more customized solutions which would be impossible from an out-of-the-box package. From the other side, companies which are anti Open Source (ie: Microsoft) sell consumer end programs which accomplish routine and general tasks. Often times these packages do not require support and because these packages are easy to re-invent and have relatively low development cost and therefore can be cloned by OSS projects. The OSS project called OpenOffice.org (branch from Suns Star Office) serves as perfect example of this phenomenon, many users have adopted OpenOffice.org as a replacement to Microsoft Office. Not only is OpenOffice.org capable of doing virtually everything Microsoft Office can but it can be deployed on a number of operating systems and also, its free (gratis and libre). It is evident that some companies have a lot to loose if many consumers switch to Open Source solutions.
Instead of sitting back and complaining how the customers are wrong for choosing alternative software, companies should be thinking of adapting to the change in consumers needs. There is money to be made from OSS, many misunderstand the meaning of "free" when its referred to Open Source software. While free does mean gratis, when it comes to Open Source software it is used as by its other meaning: libre. Being free of charge is not a requirement for a piece of software to be considered Open Source, how ever it is required that the software come with its source code and without patents or other limitations. In other words, switching the practice of a company from Proprietary software to OSS does not mean that they have to switch their money making strategies, companies can still charge per piece of software just as Novell or Sun is doing.
Following current trends it is not hard to foresee what direction Open Source is heading in. The future for OSS in server-type applications is looking bright, companies of these fields are happy to donate developer time and research money to keep the non-profit projects in motion. In his evaluation of Open Source, Fugetta, points out how IBM and Sun are avidly developing for Apache (86). Considering that Apache has a 67% market share of web servers and is still growing (Broersma), one can safely say it is not going anywhere. Apache is not alone, BINDS, mySQL and many other Open Source server-solutions are also looking strong. The future for OSS in personal and consumer level programs is also looking quite good. Although the enthusiasms by users is not as great there is a sizable number of developers willing to work on these projects. Fortunately developers are beginning to place more emphasis on user-friendliness, with this more users are likely to deploy Open Source programs to address their software needs.
- Asiri, Sami. "Open Source Software." ACM SIGCAS Computers and Society 33.1 (March 2003): 2.
- Broersma, Matthew. " Apache zooms away from Microsoft's Web server."Zdnet 12 Jan. 2004.
- Ferris, Paul. "The Age of Corporate Open Source Enlightenment." Queue 1.5 (July/August 2003): 34-44.
- Fuggeta, Alfonso. "Open source software -- An evaluation." Journal of Systems and Software 66.1 (Apil 15 2003): 77-90.
- Stallman, Richard. "Why 'Free Software' is better than 'Open Source.'"Free Software, Free Society: The Selected Essays of Richard M. Stallman. Ed: Joshua Gay. Boston:Free Software Foundation 2002
Been a while since I last posted an update. I've been mainly occupied with college, but yet I've managed to throw together some new projects. First of all I have started working on a php gallery script to dynamically make galleries using imagemagic (gdimage lacks different format support so I abandonned it during the first stage). Also I have finished a project of a different sort, installing linux on a gateway handbook 2000, a 486 sub laptop computing device. Right now its purpose is to act as tool to controll other machines from bed. I'll post pics and tech notes about it when I get a chance (I am making this entry from bed using elinks).2005-04-08 01:49:14
I hope you all enjoyed my little prank. (Server was not shutdown by MPAA after all)2005-04-01 23:46:18
Happy Solistice all.
A number of new an interesting things have unfolded on this front. First of all I recently finished my first semester in college. So huray! Now to the more geeky stuff. WE HAVE MOVED. This site has moved to new server (which I am renting together with some friends) You no longer need the :3080 at the end of the domain in the url (http://dotCOMmie.net) I have fixed some flaws that have opened up as a result of the move but I bet there is a load of broken links left. Also I have slightly updated this site, it should now be a lot more friendly to text browsers (I tested it in Elinks).2004-12-21 15:08:15
Workers of Internet boom/bust. (base on Netslaves 2.0)
During the begging of the internet boom many of the new workers thought that they had come to heaven. The field was open with opportunities, the pay was great and all the free coffee you could drink. Many of these workers came to work for start ups, they had their own reasons for doing so, some wanted a shorter climb through the corporate ladder while others wanted to cash in on the stock options. But few were really aware what a wild ride they had gotten on.
The internet industry ticked by its own clock, everything seemed to change by the hour. One day knowing HTML could land you a job with a salary of eighty thousand a year, the next you better learn to code or you'll end up on the street. But the glasses were half full and the people looked past the risk and focused on their future success. This was the spirit that was driving the internet culture.
The working conditions in these start ups were unique. There were two main concerns that were running through the head of the average employee, getting rich and keeping their job. Doing so required for each employee to keep pace, your resume better evolve at a rate faster than the internet. And one should always keep in mind that your company might not be there tomorrow, merging and bankruptcies were all too common. When entering a startup every employee knew that they would not have a regular 9-5--Monday Friday schedule. But instead they were expected to put in as many hours as needed to complete their project. For example in NetSlaves 2.0 we read of how Jan was working every day, non stop with almost no sleep and when she asked her boss if she could have a day off he replied "We're only twenty-nine days, sixteen hours and 11 minutes away from D-Day. Can't you stick it out 'till then?" It is quite ironic how in such a work centered society job security was a luxury.
Many companies went down in the internet bust, but yet some remained. While it is hard to define which set of characteristics would definitely lead to success, we do see, through NetSlaves 2.0, some practices which would most certainly lead to doom. One of the most general but reoccurring characteristics is optimism. When a company would take a slight turn for the worse both its workers and managers would just hope for the best and try to avoid the worst. For example, in the story of CRT we see how many of the insiders, especially programmers, knew that the company had no real product to back all its marketing. This fact was not a great concern to them; they continued to go to work expecting their job be there tomorrow, this feeling remained the same even when the companies great lie was to be busted. When the time came most of the optimistic workers were laid of. Only one person came out on top, as the whole fiasco had unfolded. Vincent had been long bothered by the company practice, on his free time after work he had started working on a project of his own hoping to cash in if his greatest fears were to come true. But even he did not escape unsaved. When Aether acquired CRT, they realized that they had spent millions of dollars on a worthless company that had no product. The only thing they could cash in on was the Parser, Vincent's own little jewel. In the end Vincent and Aether had broken into a series of lawsuits over the rights to the parser, leaving Vincent with a huge legal bill instead of a severance package.
Although in many other industries of the world we do see a great link between knowledge and income this link in the internet industry is very faint. Knowledge does not always guarantee a better salary, but it does guarantee better job security. In the case of a merge between two companies if you are skilled enough you might be able to get a new position at your new parent company. Income depended much more on ones class. In this hypothetical merge a person higher up on corporate ladder might not just get a job but also a nice severance package, either through stock or cold, hard money. Least fortunate were the "Aliens" which had to struggle each day to not disappoint anyone so as to not be fired and deported.
People have always looked for a way of making the big buck in no time. The internet boom was certainly no exception, millions of people jumped onto the raging bull in hopes of scoring big. But as we see the NetSlaves 2.0 many of them would be happy to walk away with just a few broken bones, of course is a hypothetical sense.2004-11-16 01:30:56