Slashdot Mirror

There are a variety of agents and monitoring tools that make up the Insight Management toolset. We've found that some of the tools are better than others.

Pretty much the only *essential* tool that's required is the cpqhealth drivers and daemons. This poll the health of the onboard systems such as fans, CPUs, disk arrays, etc, and will log to syslog when there is a fault. Unforunately, the open source lm_sensors and cpqarrayd packages don't talk to the hardware in the new G3 DL380's, so cpqhealth is your only option. You can find it on HP's support site, as part of the hpasm package for Linux... I grabbed my copy for RedHat 7.3 from here.

cpqhealth comes with pre-built modules for RedHat 7.3, 8.0 and a few other distros (SuSE for example). But I've found that even the most up to date stuff from HP's site only supports the kernels shipped on the RedHat cd, and nothing newer. Luckily, cpqhealth (part of the hpasm package) does allow you to build new modules. You'll need a compiler on the machine. Take a peek at this script: /opt/compaq/cpqhealth/custom_cpqhealth.sh - it will build a new cpqhealth RPM for you, containing the drivers and daemons necessary to log hardware faults to the syslog (as well as to take action on them).

The script will break when you first run it - it will look for the following two files:

/opt/compaq/cpqhealth/cpqasm/S10cpqasm
/opt/compaq/cpqhealth/hpuid

Both of which are missing in the most recent hpasm release. Create the S10cpqasm file yourself (it's just a startup script that gets dumped /etc/init.d - a simple touch of that file is fine for now - you can put a proper one together later), and copy hpuid from /bin (where it gets installed when you installed the hpasm RPM).

Once done, you'll have an RPM that installs the following:

two kernel modules: cpqasm.o, cpqevt.o

two daemons: /opt/compaq/cpqhealth/cpqasm/casmd
/opt/compaq/cpqhealth/cpqevt/cevtd

Make sure the kernel modules and daemons get loaded, and you'll now get warnings when a fan fails, disk in the RAID array dies, etc.

Even better - unlike the rest of the HP/Compaq Insight stuff, this doesn't use SNMP, doesn't install a web server that listens to 0.0.0.0, and seems to work quite well.

Other annoying things I've discovered about the rest of the HP/Compaq toolset:

• Dumb OS-detection routines - the Storage Monitor Agent greps /etc/issue for "RedHat" + "7.3"...
• Over-reliance on SNMP.
• Unclear documentation - the agents indicate they'll work with the snmpd shipped with RedHat, whereas HPs site indicates you need to install the HP-modded snmpd.
• Corporate schizophrenia - some daemons called hpxxxx, others called cpqxxxx ... some scripts fail, looking for the old filename (HP renamed Compaq daemons, forgot to update script).
• Check ports 2301 and 2381 - One or more of the agents installs webservers there, for "remote management". They listen to 0.0.0.0 and have no IP-filtering ability. So make sure that it's firewalled off. netstat -anp is your friend.

In the end, we just ended up installing cpqhealth on the boxes to warn us of hardware problems, and will use RRDtool for our other monitoring requirements.

You can try Cacti. It's a web interface in PHP built around MySQL and RRDtool, the much more flexible successor to the flexible-as-hell MRTG. A friend just showed it to me yesterday, and I'm playing now. Looks very very nice. Homepage for Oetikers RRDTool is here

I had a Palm with my previous employer, but had to give it back when I took my current position. They gave me an iPaq 3600 that I installed Linux on and I tried both Opie and Qtopia. While I liked all the apps, and the fact I could use Perl (even Perl/Tk), the PIM apps sucked. I struggled with syncing with Evolution, which I need to stay connected with our Corp. Exchange server.

I really missed the ease of use of my Palm, so I finally gave up and bought a Palm Pilot off ebay for $30. Instead of mucking around with a paritally useful iPaq I have what I want...Palm OS based device that just works.

Yes, I tried to get used to WindowsCE, but I think the apps suck on that platform too. Another strick against the iPaq was the battery life. It spent more time in the cradle than working for me.

Quicker

I think that internet voting should provide an improvement compared to casting paper ballots, but from there to saving the world, the expectations are quite high! Can you ensure that nobody tampers with your paper ballots? (Of course not).

My father went many times counting ballots and told me that he could recognize the vote's owner from his or her writing (out of about 500 ballots cast there). And in Ticino (southern Switzerland), all vote results are published by town: some are so small that they have only a dozen voters, so that it's quite easy to reverse engineer the vote!

I think this vote was an improvement, not because it was more secure that casting paper ballots (nor was it less secure), but because it encouraged more people to vote. 22% of the voters where regular abstainers.

-Patrik
-------------
Discover the Bluebottle OS!

With the exception of Donald Knuth, all of the names you list are of people who had mostly engineering contributions, as opposed to bringing scientific advancements in the field (although the two are somewhat related). Did you mean to exclude the people who created and formalized computer science? If not, then you most definitely want to include Alan Turing, Edsger Dijsktra, C. Antony R. Hoare, Niklaus Wirth, and Marvin Minsky.

I wouldn't doubt it. i was looking at his web site earlier, and oh my god have a lot of people bought him CDs and DVDs as gifts! he's got a list.

The configure script for mrtg claims to order CD's for the primary author a few times while beeping loudly.

You can find more information about the project on Bernt Schiele's Group Homepage at
http://www.vision.ethz.ch/projects/

Merry Xmas to all of you /.'ers

LVS was able to handle a medium-sized HTTP/HTTPS load at my last job quite well. It had 6 months of uptime serving 5-10 hits/second, and I literally never had to worry about it going down. In combination with mon, bringing machines up and down was never a problem, and failure situations were handled without the end user noticing.

Installation was a bit frustrating because I hadn't dealt with the networking issues before (the ARP problem). However, in the end it was only my lack of networking knowledge that was lacking, and the ARP problem turned out to be simple to overcome.

Support from the mailing list was great, I got thorough replies to my questions in a few hours. The documentation is good, although some parts of the HOWTO could be trimmed back a bit (more information than is needed to understand the problem, takes a bit of time to filter).

The hardware was two slower UP boxes (one live, one for failover), and the load was esstentially 0, even with mon and MRTG running.

LVS is of course just the load balancer, and the setup also included mon for monitoring, heartbeat for failover, and MRTG for trending. They all play well together, and create a very reliable, informative, load balancer setup.

Depending on your setup, one of the meta-packages such as Ultra Money or Redhat's HA suite might be best, but installing the components individually isn't much of a hassle either.

and it's still one of only a few compilable languages (excepting gcj == java) that have a gc.

There is nothing special about a "compilable language" (whatever that means) using GC. Lisp has been doing it for decades (and yes, most Lisp systems are native code compilers, such as CMUCL, Allegro, CormanLisp, SBCL, etc). Oberon-2 compilers use GC, including the open source OOC and Oberon System3 from ETH. Ada was designed such that GC could be implemented, but it rarely is. Many FP languages use GC, such as Haskell. Haskell compilers, such as GHC, NHC, and HBC all use GC.

If you haven't gotten the point yet, there is nothing special about implementing languages using garbage collection, and furthermore, there was nothing innovative when Meyer decided to use it for Eiffel.

Eifel also has plug in and compiler support for Microsoft's .NET initiative. It's pretty cool, actually, because although .NET does not have support for genericity in V1.0 (though it's planned for v2.0), Eifel.NET has overcome this limitation and allows for multiple inheritance and other cool stuff. Take a look here.


There's also an MSDN article.

-jerdenn

Packet filtering has been done for video many times.
Its an interesting variant where network filters
read frequency band tags in the headers and decide
whether to keep it or not (see for ex.
wavevideo

It's not only for storage but mainly for online
applications, it works especially well for
multicast traffic.

Adobe has been distributing SVG viewer as part of the Acrobat 5 download for over a year now.

Nobody's waiting for Microsoft to innovate SVG or do their XDocs whatever thing; check these static examples generated from MS apps with SVGmaker: Excel, PowerPoint, Word, Project

For building SVG web applications its true that there aren't comprehensive IDE tools available yet, but that hasn't stopped developers from creating some definitive web apps with simple home grown tools (starting with a text editor since SVG is just XML).

Like this interactive logical diagram

Check this awesome mapping example

And this wonderful airport flight management app.

I'm convinced - there is no need for more than 5 computers, worldwide. (Thomas J. Watson, Chairman IBM, 1943)

I mean, I know IBM believes in low risk, but sticking to a 60 year old business plan is really pushing the issue. ;-)

One of my favorites is the sodium acetate tower. It is a very safe demo that gets a good reaction out of just about any age group. You make a supersaturated solution of sodium acetate in a spotlessly clean beaker. Show everyone the clear liquid (looks like water) then start slowly pouring it on the table. Crystals of sodium acetate form as you pour, and the water is trapped within the crystals. You wind up with a pile of fairly dry looking sodium acetate and no liquid. Very impressive. Sodium Acetate Tower

Another one I like is the burning carbon disulfide demo. I've seen this done using a long glass tube full of carbon disulfide gas. Drop a glowing splint in one end of the tube, and as it falls you get an amazing blue flame. Here's a link (hope you speak a little German) CS2

They did it a bit differently. As you might guess, this lab is a bit more hazardous and you do get some stink from the sulfur. It's pretty though.

Making your own mirror is another great demo. You prepare a small batch of silvering solution. ISTR using silver nitrate and nitric acid, maybe using an aldehyde as a reducing agent. I'll try to link to a recipe. Anyway, you mix the solution in a round bottom flask and begin swiriling. It takes about a minute, but as you swirl a silver mirror plates out onto the glass. Tollens Mirror

I used a bit of a different procedure, but this looks like it should work. You may consider keeping the flasks a little on the warm side (100-120 F) just before you do the demo. I've gotten better results compared to using cold glassware.

A great set of books is Tested Chemical Demonstrations, Vol. 1-4, by B. Shakishiri (University of Wisconsin Press.)

Power Mac G4 Toaster

Where did you check?

"Francium's most stable isotope, francium-223, has a half-life of about 22 minutes. It decays into radium-223 through beta decay or into astatine-219 through alpha decay." source

In all three of the most stable isotopes, alpha decay is less likely than beta decay for the first step, and if you look at the decay trees here, you'll see that you are pretty much guaranteed to get some beta decay someplace along the line.

MRTG. Need I say more?

haven't seen the Unified Modeling Language used much in the OpenSource enviroments where UML will be used

In some ways, that's a pretty severe indictment of Open Source. UML is about software analysis and design.

You don't quite come out and say it, but I'm getting the impression of

OpenSource doesn't use UML.
UML is about software analysis and design.
Therefore, OpenSource doesn't do software analysis and design.

That is to say, UML is not the only way to do software analysis and design. Bertrand Meyer has thrashed UML pretty hard: So, although UML willbe successful at first, because it has the right endorsements, it will be of little use to the actual process of developing software. (Unfortunately, I can't find the article where he let loose with both barrels.)

most Open Source hackers don't have the patience to design their software

Or the knowledge - I don't remember anything in any of my compsci classes about designing software. Open Source ad-hoc design, build, and redesign and rebuild seems to have worked pretty well in some cases.

or validate it

Money, perhaps? Spending several thousand dollars for a validation kit that I can only use on my machine is quite pricy.

or write documentation, or do usability studies

Serious user documentation in the commericial world is done by writers, not programmers. You can't expect people whose skills are programming to do everything.

If you're interested in real electronic voting (not just replacing the punch card with a keyboard in the voting booth) I suggest you start reading here.

Open source is not the solution. Good crypto is.

-jfedor

• First of all, this is a PhD project, not a product in development.
• Second, use your imaginations - just because it's demonstrated on Ikea furniture doesn't restrict it to that. Ikea furniture is designed to be
  1. compact to ship, and
  2. easy to assemble.
  Having some sort of assembly assistance could relax the latter requirement and enable more complicated DIY products. Or they could be used for faster training of assembly-line workers. If I get any more ideas, I may have to start charging consulting fees.

On to my subject: There's a 116 MB video on their site. I downloaded the whole thing at work (way fast) and watched it (about 5 minutes or so). It's pretty deadpan, and shows a guy putting together an Ikea Pax armoire unit. (It just so happens that I have three of these myself. They're pretty straight-forward to assemble, just quite heavy at 50 kilos per unit, not including doors or shelves.) There's also footage of the developer discussing how his ideas work, with some overlays of accelerometer output and the like. The clip ends with the builder standing proudly next to the completed armoire, as the image fades to black. After a short pause, there is a loud crash, so I think these guys had a sense of humor about their project.

For those with maths, Messers Wolfram tell all. I like this one.

This experiment shows off wave/partical duality (it even has cool terminology). The cool bit about physics (yeah, it has cool bits) is the things it takes your head a while to get around.

OK, background: waves spread round corners. Think of a wave at a harbour mouth. The closer the gap is to the wavelength of the wave, the better it spreads (look up diffraction) (troll me, I know this is a gross over-simplification) - ever think about how you can hear but not see round corners? Light == really short wavelengths (nanometres), not like door width lengths (m) (doesn't bend well round the corner), sound == long wavelengths, kinda door-width like (m/cm ish) (bends very well round the corner).

So you get two bits of card with a light behind them, and a screen to shine light through them onto. The first card has one slit, so it shines a little line of light onto the second.

The second has two parallel slits in it, within range of the spread of light, and the light that gets through the first card onto a slit in the second card makes it to the screen.

Now the cool bit.

You get a ripple of light on the screen. Not a black screen. Not two lines showing up the second card shape. Ripples.

Now, modern physics can explain this. It's the wavefront from the first slit (think ripple hitting a harbour mouth) that spreads out in a circle and hits the next two slots, starting another ripple on the other side of both.

At the far wall, you get points where the peak of a wave from one slit hits the peak of a wave from the other, and you get a really tall peak. Or a trough and a trough, and get a really low trough.

MRTG

No, PNG is not "perfectly fine" for displaying graphs on the web.

Is that why most perl-based graphics routine-using programs (namely MRTG and RRDTool) use PNG as their graphing format?

No, PNG is not "perfectly fine" for displaying graphs on the web.

Is that why most perl-based graphics routine-using programs (namely MRTG and RRDTool) use PNG as their graphing format?

In a sense, yes. In nature, the 'robot' is an animal, and it evolves its body at the same time as evolving the 'circuitry' in its brain to control the body (given one is a subset of the other). There really is no difference between a GA that finds an optimal physical configuration (i.e. the robot's body) and a GA that find an optimal way of controlling that physical configuration. The researchers decided to focus on the latter question (i.e. for simplicity; this doesn't invalidate their efforts). I'm sure you'd find stuff in the GA literature about using GAs to design physical entities (I like to use the "Designing a Training Shoe" analogy when explaining GAs to people).

>as even we humans learned to assist the limitations of your body

Yes we did, through evolving a highly sophisticated brain (it says here...), we can build wrist watches to help us tell the time etc. Again, a GA mechanism at play.

>Trial and error is an excellent learning tool, look at how much toddlers rely on it... I cry I get food, etc.

I don't know too much about child psychology, but the machine-learning model of this is the Neural Network. Associations that are useful get strengthened (and hence get a chance of becoming part of the solution), whilst those that are not die off -- so we have a "survival of the fittest" situation again.

Ah, thank Google. I think it was Oberon.

The example file he provides is quite interesting - there seem to be three major dependency points; fileutils (which you'd expect), perl (ditto), and python (huh?).

I guess the python dependency comes from some of the configuration tools that Red Hat includes - can anyone confirm that?

I dunno what exactly I'd do if someone DID start leaching, since I have no real contract, but then again, I have the switch in my condo, so all I need to do is pull the plug.

The Linux traffic control stuff is amazing! See Advanced Routing and Traffic Control HOWTO for details on how to make a router that is good about prioritizing traffic. If you have a leech, you could just put his traffic as lowest priority; he could suck all the bits he wants with minimal problems for others.

Also, the social solution can work pretty well. put together some MRTGgraphs of who's using what; then when people gripe, show them where to check who's using what. That way, you aren't the bad cop.

As to the question of flashing the LED because Morse on the speaker might be too annoying, I say go for the speaker.

Point-based rendering has shown some amazing results -- QSplat, for example, provides results in realtime that are flat out unimaginable out of traditional engines. Even higher quality output is coming out of the Surface splatting hackers.

Image based systems also seem to be yielding results -- Gondry's Star Guitar video, which showed scenes from a window of a train synchronized to music, was undeniably compelling and could simply not have been done with traditional 3D approaches. Schodel and Essa's work with Video Sprites are also quite impressive.

I don't mean to provide a litany of unusual rendering techniques for you to ponder. I bring them up because polygonal approaches have clearly yielded some incredible results, and I'm interested to know whether you think point-based and/or image-based strategies will yield similarly disruptive fruit. Also, I'm curious whether you're aware of any other particularly obscure but powerful methods for scene generation.

So, in short: What's next for 3D?

Yours Truly,

Dan Kaminsky
DoxPara Research
http://www.doxpara.com

My part-time project/dream is a modern re-implementation, where the TeX typesetting algorithms are embedded in a modern (Common Lisp) environment--so you can code TeX formats and macros in a heavy-duty honest-to-god programming language, and have an high-level, truly modular implementation using real data structures that could actually be tweaked and modified to do even funkier typesetting tasks.

http://www.cs.inf.ethz.ch/~skoupy/
http://www.ntg.nl/eurotex/skoupy.pdf

check out his other books,Things a Computer Scientist Rarely Talks About [amazon.com], and 3: 16: Bible Texts Illuminated [amazon.com], for explanations of why not every christian is a fool.

You need to study some geography.

You should have posted a link to MRTG so I didn't have to tell Google that I was feeling lucky! :-)

Wow. I've seen graphs like this before, but didn't know what the programs were called. I always thought they were from Apache. I'll have to check out MRTG when I get home this evening. Thanks for mentioning it.

You should have posted a link to MRTG so I didn't have to tell Google that I was feeling lucky! :-)

Wow. I've seen graphs like this before, but didn't know what the programs were called. I always thought they were from Apache. I'll have to check out MRTG when I get home this evening. Thanks for mentioning it.

I currently have my joystick measuring the temperature of my room once a minute. I originally planned to use thermistors, but experiments with a diode turned out to work just fine. Now, I honestly have no clue how it works, because the specific diode I'm using is basically broken (doesn't act like a diode should in a normal circuit), and a normal diode doesn't work in my circuit. So it was just complete luck that it worked out for me (and I *really* don't know why it works...) But it should work theoretically with thermistors in the same way. Essentially, as long as your current varies with temperature, you can measure temperature with it.

The code is amazingly simple. Here's the important part (C, obviously, running on a Linux 2.2 kernel):

#include <sys/time.h>
#include <asm/io.h>

long getreading(short int which) {
struct timeval before,after;

outb(1,0x201); // Poke the monostable multivibrator ;)

gettimeofday(&before,NULL);
while( (inb(0x201) >> which) & 0x1 );
gettimeofday(&after,NULL);

if(after.tv_sec > before.tv_sec) after.tv_usec += 1000000;

return (after.tv_usec - before.tv_usec);
}

"which" is just the channel you want to read.

The tricky part is the calibration. You'll first just hope that your readings scale linearly with temperature, and mess around with an offset and slope until it matches some known readings. If it's not linear, well... But if you just want "Good", "Bad", and "This reading wasn't taken, because the CPU is a puddle of slag", then it's not so bad.

Good luck. It was a fun project for me. I still don't have the calibration worked out quite right, but that's okay. Oh, and I use RRD Tool to graph the results. I have pretty, colorful, utterly useless graphs of the temperature of my room. Yay!

To give a real life example, I did some work at one site recently who wanted some very basic network usage monitoring - the kind of stuff that would be a no brainer for MRTG. Because of the KISS argument above they were about to spend the best part of £1000 on an overkill commerical product, with additional outlay as they added additional devices because it was licensed per node. They were happy with this because they understood the license and costs, even when I said I could set up an a Linux box with an MRTG config for then in a morning and have a web configuration front-end tailored for them by the end of the same day.

(For those who want to know - they agreed to give it a shot, and I was in the local pub with it all done by 14:00 the following day. They were so happy and their license concerns sufficiently assuaged that I'm replacing their NT based mail gateway with a Linux box next month...)

None of these are X11 alternatives on the level of SVGALIB or DirectFB, but a bit higher level. They require a low-level display medium like DirectFB, SDL, or X11 (but you can ignore that option for now).

Squeak Smalltalk: A cool Smalltalk environment. Based on Smalltalk-80, for which first modern WIMP was invented. Has a bunch of little apps, simple web browser, vt100 client, few email clients, web servers, a couple different GUI toolkits and programming paradigms to choose from. Personally, what I use mostly as my OS. I like having my entire environment available to me, to be changed as I like, in a very straightforward way. Rather like Emacs users, I suppose. Except Squeak is more customizable, and has full windowing system. Also can run as the OS, no Linux or X11. DirectFB, SDL, X11, Mac (9/X), Windows, Acorn, WinCE, BeOS and lots of other ports that all run the same binaries.

ETH Oberon: Implementation of the Oberon language - derived from Pascal and Modula, by Nick Wirth. Has it's own entire GUI system, like Squeak does. Can run as an OS, without Linux or X11. Also has a VNC client, so you could still run the X11 app or two that you still needed in a window. :)

PicoGUI: A really cool GUI system especially for PDAs and other embedded applications. Super fast. Bindings for C, Perl, and Python (I think). Linux FB and SDL ports, runs wherever they can. Not much in the way of apps thus far, but it's definitely alive and under pretty active development.

QT/Embedded: You know, like runs on the Zaurus.

GTK+ on Direct FB: Can't say I've used this, but I imagine bindings for regular GTK+ work in this port, which makes for a lot of development options.

MicroWindows/Nano-X: Yet enother embedded GUI option. It's developer seems to be pushing for PDA, set-tops and such. Not many apps, but could be useful especially for custom apps.

Are there any worthwhile just-Java windowing systems out there? There are al ot of Java-OS projects, but none of them seem to have gotten past linking Kaffee with OSKit...

Probably others out there, but this is a good look at some options.

Have a look at CWEB by D.E.Knuth. It shows a way of writing C / C++ / Java code that can really be understood.
Apart from that, the use of syntactically more rigid programming languages like Oberon reduce the chance of obfuscated code in the beginning.

Actually, that is often not true at all. Speaking as a grad student with experience in structural biology, the majority of programs actually used for NMR structures, X-ray structures, , molecular graphics, etc etc (the list it very long!) are all developed by university labs and given away free, generally open-sourced as well. Universities don't generally hold such things hostage, as there is the understanding that science is based on sharing, nor hoarding.

-Ted

One of the most common problems I've encountered in my years as a systems administrator is poorly managed networks. If a network is designed without the presence of mind anticpating DoS attacks, then frankly, the victim company deserves *some* of the blame for the problem.

One mid-sized ISP I worked for had been operating for 5 years prior to my employ and the network operators had never heard of monitoring tools like MRTG, RRDTool, Netsaint or Big Brother etc etc!

"We do it to ourselves and that's what really hurts" -- Radio Head.

-- Steve.

One of the most common problems I've encountered in my years as a systems administrator is poorly managed networks. If a network is designed without the presence of mind anticpating DoS attacks, then frankly, the victim company deserves *some* of the blame for the problem.

One mid-sized ISP I worked for had been operating for 5 years prior to my employ and the network operators had never heard of monitoring tools like MRTG, RRDTool, Netsaint or Big Brother etc etc!

"We do it to ourselves and that's what really hurts" -- Radio Head.

-- Steve.

Damian Conway will give a public talk at ETH Zurich (Swiss Federal Institute of Technology) on Monday, February 11th, 2002 at 17:00h.Everybody is welcome. Location: Departement of Information Technology and Electrical engineering, Gloriastrasse 35; Auditorium E6. http://www.ee.ethz.ch/events/index.en.html Abstract

Damian Conway will give a public talk at ETH Zurich (Swiss Federal Institute of Technology) on Monday, February 11th, 2002 at 17:00h.Everybody is welcome. Location: Departement of Information Technology and Electrical engineering, Gloriastrasse 35; Auditorium E6. http://www.ee.ethz.ch/events/index.en.html Abstract

I've recently set up a master-master replication environnement on Oracle 9i and I did some research to check if it was possible with postgres.

In fact there are many solutions available (check techdocs.postgresql.org for a list...)

The most advanced guys on the subject seem to be the swiss engineering school in Zürich. Here is a list of their publications.
They seem to have developped a replication scheme (Postgres-R) where they have better than linear performance improvement when they add new masters...Quite impressive

Quentin

If you are creating large complex documents you really should take a look at LaTeX. All it takes is 95 minutes.

This does most of that.

Tiling windows, which is how I always arrange my windows in practice anyway, have been around at least since the late eighties in Niklaus Wirth's original Oberon system. I'm not sure what it looks like now that others have gone jumping on bandwagons since Wirth's retirement. (here and here)

Anyone with Oberon experience out there?? It started out as a tiled windows system only, but now they've developed an overlapping windows desktop as well. Checkout the screenshots.

Their comment on tiled display is useful: The Gadgets desktop also has a tiled display mode with two vertical tracks. In this mode a newly opened viewer automatically covers half of the largest existing viewer in the track. This is ideal for text-based work, e.g., programming or text editing. Viewers can be resized vertically and moved, but they always use the full track width. Because there are fewer degrees of freedom, it is much quicker to arrange viewers optimally. newly opened viewer automatically covers half of the largest existing viewer in the track. BTW, windows are called viewers in Oberon.