Huh? What?
An E-playground for members of Bebear.net
Subscribe
Receive the RSS Feed of Samsara updates.
Links
People's pages that live on this server
Mon, 20 Nov 2006
Synergy clipboard on OSX busts if started via SSH
I've been using synergy2 to help manage working with multiple PCs simultaneously. It's a great tool but sometimes strange errors can happen.
The problem
One error I finally decided to track down is trying to start a synergy client from a SSH connection in OSX to a synergy server. While the mouse and keyboard works. The clipboard functionality completely fails. What this means is you can't cut and paste from the machine running the synergy server to the OSX machine.
The workaround
The workaround is not too hard but a little annoying. You MUST start the synergy client locally on the OS X machine rather than SSH into the OS X machine. Perhaps when you SSH in, the synergy client isn't able to get a hook on the correct WindowServer or something so clipboard functionality goes south. However mouse and keyboard still work.... strange.
References:
Synergy 2's Bug Tracking database
Synergy 2
Why compatibility is important for large corps
Their operations division is responsible for 9,000 different install scripts for their employees around the world.
That's INSTALL scripts for programs. Not the programs themselves. I think you can do the math from there.
The first international workshop on Plan 9
If you are in Madrid, Spain during the Dec 4-5th timeframe of 2006 it
looks like you can jump in on the first
Plan 9 international workshop.
I'm still trying to get a machine that runs Plan 9 without much fuss to see how it is but I always find new information on Plan 9 interesting to read.
Plan 9 Workshop site
OSNews link to Plan 9 workshop
via James Duncan Davidson's blog
Tue, 14 Nov 2006
And what does he have to say? (Well at least through the reporter filter)
Kihara has written in his books that it still boiled his blood to think that
consumers have been forced to use the "inferior" VHS over Sony's Betamax.
Sony has had a history of good engineering quality. However, lately they've gotten soft and in my opinion really started NOT getting it when it comes to engineering the next tool. Example, The Mylo.
Sony really needs to wake up and realize the world of electronics is a much larger place than the little playground they had to themselves for decades and start learning to play ALONG with others rather than create more messes that takes years to resolve.
As for 'inferior' VHS. Well, if Apple wasn't stupid back in the 80s Microsoft wouldn't exist. 'nuff said
Anything that means solar technology is getting cheaper is a good thing in my book although the link is to a Technology Review article. So far Technology Review keeps talking about great technologies that I rarely see in my daily existence. But they do make great press. But I can always be hopeful.
The Technology Review article via Slashdot
The OLPC has won an award already but no one is actually USING one
This really gets to me. The OLPC has
won a
Popular Science award
yet it isn't even out yet. I'm sorry 'in manufacturing' does NOT mean it's
in the hands of children who need it most. Until I see THAT I watch by
the sidelines skeptical.
Here's something you probably won't read in the local Japanese newspapers:
This is the true face of the PlayStation 3 debut in Japan. Hardcore gamers are not here waiting in line overnight, buying a first-run PS3, and running home to play some good old next-gen gaming. Rather, opportunistic Japanese businessmen have the largest presence, hiring poor Chinese men and women to wait in line for a PS3
Frankly, this is yet another blunder for Sony but I bet it will be silenced by the Japanese media so the only people who will know about it are:
- People reading this blog (All 3 of you!)
- Gamers not in Japan
- All the Kotaku.com readers
Picasa and Gift CD update (Beware the Vmware)
Well, after that really long rant I tried burning a DVD-R with some plain burning software and that refused to read under this drive as well. I finally figured out what the problem was. It had nothing to do with Picasa at all. It had everything to do with Vmware grabbing the DVD-R and sending it to a virtual instance I had running. *sigh* I hate when that happens. Well, I guess I should feel sheepish giving Google hate mail now. Ah well...
Picasa Gift CD Problems on version 2.5.0
I've been using Google's Picasa
photo management tool for some time under Windows and have been satisfied
with the very low layer substrate that is used to manage photos (folders
correspond to folders in the filesystem). I updated to version 2.5.0
some time ago and tried burning a Gift CD and noticed I had a couple of
coasters after this point. After searching the forums for a little bit
I found
this:
I have maybe a clue for this problem. It seems that de the Gift CD
feature stops working if you install the latest version of the
software.
Conclusion? The latest software breaks functionality but adds lots of eye candy. Sorry Google, this is another strike for you guys. I really like Picasa since it sucks less than most of the other solutions for photo management so far (including iPhoto) but finding out stupid stuff like this doesn't enamor me to Google's fabled Google Way of doing things. Who decided to make more eye candy and take away functionality? Frankly I don't give a crap about Blog This (I don't use Blogger) / Order Prints / Using this stupid Hello service / Tivo Export. But I DO want a Gift CD so I can hand it to anyone. Downgrade here I come.
Google's Support Forum on Gift CD becoming sucky:
Sat, 11 Nov 2006Apress released a book titled Practical Ocaml and I was excited about the book and read the interview with the author. However, after looking briefly through the comments on Lambda the Ultimate I might have to reconsider purchasing this book.
Ehud
comments:
I haven't seen the book yet, but it's really sad if the book is as problematic as this thread indicates. I had high hopes for the "Practical X" line of books. I thought it might become an O'reilly-like brand for books about non-mainstream language
Matt comments
:
This book is a big disappointment. As already noted, it has some severe stylistic problems and the writing is uninspiring, to say the least. But by far the worst problem is that the author's knowledge of his subject is simply insufficient. This shows up all over the place, but I'll highlight two examples that I think exemplify the issue.
William comments
:
Some items are given only a superficial introduction (e.g. variants -- a very important part of OCaml), and then used fairly heavily two chapters later without so much as a pointer back to their introduction for people skimming the book.
and finally
Merjis (one of the technical reviewers)
writes
on his blog :
I only wrote one candid assessment, and decency and privacy prevent me from
disclosing what was in it. I will just say that it was not positive.
Needless to say, a very disappointing round of comments. The part regarding poor introduction of material then abusing it later on is a serious flaw in my assessment of the book. I have been suffering this problem for years on many different topics (maybe it's just because I'm brain damaged) and it's one of the major things I always am critical about when evaluating the learning process on anything. Crappy treatment of the foundations you are trying to build the rest of your mental model on is terrible.
Related
- A Lambda the Ultimate contributers view on Ocaml the language itself is available here
- A much better (and free) introduction to Ocaml (supposedly) is available here
Photosynth, a new way of seeing relations in your photo collection
I just took a glance at photosynch and am really impressed with the applications of computer vision and image processing techniques to create a really unique application.
The basic idea is to take a pile of photos that are related to each other somehow (imagine taking zillions of pictures of the Taj Mahal from tons of different places) find similar features in all the images and try to reconstruct a mock 3d space that shows the spacial relation between all of your photos. This is really cool as you might be able to create a very interesting photo tour from your photo collection in a 3d navigatable space.
Oddly, I was trying to come up with a similar idea to link videos stills in QuickTimeVR movies and try to use the linkable features in QuicktimeVR to provide clickable hotspots that would take you to another photo that was a picture of the same scene however this is far slicker and if it works with very little intervention from the user besides pointing to a pile of photos and letting it do its job that would be great.
However, there are still caveats. The whole process takes hours or days to currently do and the current technology preview is only for a pre-rendered project. The true acid test will be in my opinion the ability to just point to a folder of pictures and have it do its job with as little possible human intervention as possible. That is a not a trivial problem but I'm sure we'll see something interesting especially since it has two (very well) known researchers in the computer vision field. I'm really looking forward to the results of their labor. My last question is how many technologies behind this are patented already. It'd be great if it an OSS implementation inspired from this project could be made however patents are a sticky problem.
Try it yourself
(
6 or 7 REQUIRED)
I noticed Charles Nutter (one of the JRuby developers) mentioning that the fork() method in JRuby will most not likely be supported:
We would strongly prefer to avoid any implementation that requires fork, since we can't really support fork in JRuby.
While I can understand the difficulties in the Java VM giving lots of hell trying to handle something like fork() I feel it's a shame since it means if you use fork and quite a few process related calls in Ruby it will not be portable in JRuby. For a sysadmin, handling processes is one of their jobs so personally, I hope they can come up with some way to handle it somehow. But JRuby's strengths might not be in a scripting language for sys admins but providing a way to integrate rails with the monster that is J2EE. How cool would it be to develop a rails application all in the context of a java web application that you can just deploy right into a Java application server? I think this could be JRuby's little nitch (besides being an integration tool for java devs)
THE Official 8800GTX Performance Comparison
This is awesome. I won't ruin it for you by posting more than the link.
Check it out
Chumby, an open hardware flash player
Lots of discussion on Chumby has been happening (
here,
here,
here and
here).
A couple of points on Chumby are:
- An open source hardware flash player
- Schematics completely open
- Runs Linux
- The hacker that wrote Hacking the Xbox is one of the leads behind this
- Lots of O'Reilly alpha geeks seem utterly into this
- Price point at around US$150
- A reverse-engineered open set of creative tools
- A more open specification than the one that is available
- Adobe opening up their toolsets
If normal biff needs some more shinyness to it..
cdbiff execute `eject' command to eject a CD-ROM tray when mail arrives.
Fixing 'mkmf' load error Ruby in Ubuntu
If you're a Debian or Ubuntu user you'll find that the Ruby standard distribution is split into lots of little packages so doing something like apt-get install ruby only gives you the ruby binary and a subset of the libraries for Ruby. You'll need to add more packages if you want to utilize more of Ruby's standard library. This is all in order to comply with Debian packaging Guidelines. That is all well and good but a little frustrating when you 'just want to do work'.
For example, I wanted to install
sys-proctable
(Something I really believe should be in Ruby's standard distribution)
however ran into the following:
$ sudo gem install sys-proctable
Building native extensions. This could take a while...
extconf.rb:8:in `require': no such file to load -- mkmf (LoadError)
from extconf.rb:8
ERROR: While executing gem ... (RuntimeError)
ERROR: Failed to build gem native extension.
Gem files will remain installed in /usr/lib/ruby/gems/1.8/gems/sys-proctable-0.7.3 for inspection.
To get around this you will need to install ruby1.8-dev (or ruby1.6-dev or whatever version of ruby you're running) to get mkmf.rb which is listed in the Ruby Standard Library. I'm sure most Ruby hackers worth their salt consider this a 'duh' thing but not of all of us can be shining stars in the Ruby community
How I fixed it...
$ sudo apt-get install ruby1.8-dev
..
$ sudo gem install sys-proctable
Need to update 4 gems from http://gems.rubyforge.org
....
complete
Building native extensions. This could take a while...
ruby extconf.rb install sys-proctable
checking for rb_pid_t... yes
creating Makefile
make
gcc -fPIC -Wall -g -O2 -fPIC -I. -I/usr/lib/ruby/1.8/i486-linux -I/usr/lib/ruby/1.8/i486-linux -I. -DHAVE_TYPE_RB_PID_T -c proctable.c
gcc -shared -L"/usr/lib" -o proctable.so proctable.o -lruby1.8 -lpthread -ldl -lcrypt -lm -lc
make install
mkdir -p /usr/lib/ruby/gems/1.8/gems/sys-proctable-0.7.3/lib/sys
/usr/bin/install -c -m 0755 proctable.so /usr/lib/ruby/gems/1.8/gems/sys-proctable-0.7.3/lib/sys
make clean
Successfully installed sys-proctable-0.7.3
Installing ri documentation for sys-proctable-0.7.3...
Installing RDoc documentation for sys-proctable-0.7.3...
The Pied Piper of Architects (aka be careful of the wrong brain drain)
Interesting article on trying to keep your technical architects around as they are the ones that have the scars from war stories and have built up experience on leading the project in the right direction
But cynically speaking, what company is ever going to see the wisdom in that. Right? I mean keeping an old fart around on the payroll that might save you a significant chunk of change is just too forward looking...
David Pogue reviews the Sony Mylo
From the article...
Sony may be the first company ever to depict throwing up as a way to
sell electronics.
Seems once again Sony missed the target market by a decent amount again. One of these days they'll hit the right combination again. But it seems as usual, their engineers are a little out of touch with the target market.
Sat, 14 Oct 2006
Getting cvsync running on cygwin
I'm trying to get cvsync running on cygwin and after looking at the homepage. There didn't seem to be that many straightforward instructions nor is there a package in the default cygwin repository. Bummer.
So I took a shot and tried downloading the tarball and compiling it.
$ wget ftp://ftp.cvsync.org/pub/cvsync/cvsync-0.24.19.tar.gz
$ tar xvzf cvsync-0.24.19.tar.bz
$ cd cvsync-0.24.19
$ make
It blew up on a variable called HASH_TYPE in the makefile. Seems that cygwin doesn't have a native hash library to use. Luckily it spit out that it could use some hash types if fed one on a the command line. After looking at the choices I decided to use the OpenSSL one. But make sure to install openssl-devel from cygwin. Or else it won't compile. After that the compilation will go fine.
# Make sure the man directory exists or else the install barfs $ mkdir -p /usr/local/man/man1 $ HASH_TYPE=openssl make install make[1]: Entering directory `/cygdrive/c/someuser/docs/src/cvsync-0.24.19/cvscan' *** Build parameters: CC_TYPE= default CFLAGS_OPTS= LDFLAGS_OPTS= PREFIX= /usr/local ZLIB_PREFIX= USE_INET6= USE_POLL= HASH_TYPE= openssl HASH_PREFIX= /usr/local PTHREAD_TYPE= native PTHREAD_PREFIX= /usr/local SOCKS5_TYPE= SOCKS5_PREFIX= ... *Lots deleted* ...Wed, 11 Oct 2006
Google Groups Beta doesn't support page up/down or keyboard shortcuts?
I've been testing the new Google Groups Beta and design-wise it looks a little bit cleaner however one SERIOUS flaw is that I can't seem to get paging to work at all. Pressing Page Up or Page Down results in no response.
Okay, rtfm... Piles and piles of pages of useless docs on how to create and add content. But how about people who need to BROWSE content? No documentation whatsoever if Google Groups Beta uses some sort of special hotkeys for navigation. And no, I'm not going to experiment/read the JS source/use intuition to guess what they are. This is something I consider BASIC for any type of news reader tool.
And yes, I've already sent the folks at Google an email about it. However, frankly I'm a bit disappointed at the Google Overlords for missing such a brain-dead simple feature (or the failure to document it). How many advanced algorithms and UI pains did you guys go through just to forget the lowly Page Down key?
Sat, 07 Oct 2006
A social psychology model answer to why people won't switch OSes
OSNews has an interesting article detailing why people won't switch OSes even when there might be many technical advantages to. Definitely an interesting read. It frames the reasoning in something called the Elaboration Likelihood Model. I've not heard of it but then again I'm no psychologist researcher either so I guess that's not surprising.
Wed, 04 Oct 2006
Fixing SVK 'Oh no, no more exceptions! add_directory() failed'
The Problem
$ svk mirror http://svn.somewhere.org/project //mirror/project
Committed revision 1.
$ svk sync //mirror/project
Syncing http://svn.somewhere.org/project
Retrieving log information from 1 to 34
Oh no, no more exceptions! add_directory() failed. at /usr/local/share/perl5/SVN/Mirror/Ra.pm line 1044.
If you're getting the above problem happening when trying to run SVK to mirror a repository. Congratulations! You hit a bug. After searching around on Google the closest thing I found to explaining the problem is this bug.
The cause
I'm not sure what the reason but in general the problem seems to be isolated to the SVN::Mirror perl module. Perhaps if you upgrade SVK and all its dependencies via your package manager of choice and SVN::Mirror is updated to the right version I think you can get away from the problem.
A Fix
Upgrade SVN::Mirror (somehow). Since I installed SVK by hand it was possible
to upgrade using cpan. Here's what I did:
$ /usr/local/bin/cpan
cpan> install SVN::Mirror
Running install for module SVN::Mirror
Running make for C/CL/CLKAO/SVN-Mirror-0.71.tar.gz
...
...
After that, I was able to use svk sync as expected.
Well perhaps not the Ubuntu I associate with but close enough...
In Praise of the Ordinary Life
The OLPC Project finally starts showing off an implementation of their display
Can't wait to get my hands on one (million) of them.
AIBO creator's new robot is powered by NetBSD
Very cool. Who says NetBSD is
dying?
Check the specs
at the bottom and you'll see NetBSD listed..
Thanks Engadget
Mythtv + Mysql5 + Ubuntu Dapper = Busted
A little history...
I took another shot at getting mythtv setup again after trying it a few years ago and lashing myself a few times trying to get the ivtv driver patched into a mainline kernel. After a bit of work back then I got it working but times changed and I ended up using the machine for different purposes.
Installing MythTV again...
Once again, I've decided to take another shot at a mythtv installation. This time I figured I'd piggyback onto a distribution that probably had packages for it available. My choice was Ubuntu Dapper. So I did the simple thing and tried:
sudo apt-get install mythtv
Trying to configure MythTV
And soon I had a mythtv installation almost setup. It seems that many of the Japanese patches have been merged into mythtv (Wow!) so I didn't have to crawl all over the Japanese web space to figure out how to patch in Japanese patches to obscure_utility (TM) just to get this working. The first time I tried populating mythtv's database with TV listing information I got errors siilar to:
Malformed UTF-8 character (1 byte, need 3, after start byte 0xe7) at
/usr/lib/perl5/vendor_perl/5.8.8/Date/Manip.pm line 7167.
Since I'm not a perl god nor am I intimately familiar with mythtv or any of its tools, I ended up searching on that error message and found a post on a Japanese board basically saying this would not be a major problem with using mythtv in Japanese so I did what most people do... ignored it.
After I got the base configuration done, I realized I'd be spending a lot of time not looking directly at this machine since it's main job was to be a backend recording machine rather than a frontend so I installed mythweb to help me manage recordings:
sudo apt-get install mythweb
The Killer Problem settles in
This is where problems started coming in. After populating the TV listings database, mythweb still failed to show ANY listings whatsoever. Needless to say, this makes it a little hard to schedule watching anything. After waiting a few days to see if this was just a transient problem I still had no TV listings so I decided to search on it. What I found basically was Mysql 5 changed a keyword 'repeat' to a reserved word which OF COURSE mythtv relied on using in one of its tables. What this results in is a non-working setup for viewing TV listings on mythtv. This problem affects mythtv 0.18.x and the mythtv developers have basically said 'upgrade' if you want to fix it. However the Ubuntu Dapper packages stay steady at 0.18.x with a dependency on mysql-server which defaults to Mysql5. It seems there are updated packages in Edgy (Why the f' is it always in the NEXT release rather than backported to THIS release????) but none for Dapper yet. I hope someone backports these packages. In the meantime, I'll try to see if the updated packages build cleanly on Dapper for my own purposes.
References
- All bugs logged against mythtv in Ubuntu
- One bug logged against Ubuntu mythtv
- Ubuntu source packages for updated Mythtv (0.20)
- The REVU page for the MythTV package (A link on where to download all the source files for personal testing would be VERY helpful)
- Mythtv's trouble ticket on this problem
- Installing Mythtv via SVN on Ubuntu (Not recommended since 0.20 of mythtv is alread released)
- Japanese instructions for setting up Myth TV
Why evil is only a matter of time
Here is one explanation why Google will eventually become evil...
companies are at different stages of a standard software business model, which goes like this:
Getting Linux running on a Mac Pro
I've been wondering how hard it would be to get Linux running on a Mac Pro. Seems quite possible. This has a bit of promise for other things...
Bin-false's version
Ubuntu Forums on Linux for Mac Pro
Core 2 Duo Motherboards and Linux instability fun
I built myself a Core 2 Duo system recently with the hopes of running a much faster Linux setup than my old Athlon setup. However, what I ran fast into were compatibility issues. It seems that the JMicron chipset which is present in the MSI P965 Neo motherboard that I own is the culprit of many problems [1], [2], [3], [4]
A workaround which is to install Gentoo (Ubuntu Dapper folks does NOT work at this time) or some other distro besides Ubuntu and make sure to boot sending the kernel parameters all-generic-ide and irqpoll to make sure that the kernel does not completely bork itself on bootup. After that, you STILL have an issue with the Gigabit NIC that is included onboard. And NO, it is NOT in 2.6.17.x kernels or less. In fact you have to download the stupid thing from Realtek's website (HELLO, have we heard of merging into the kernel??) which can be found here.
Just for note, the 2.6.17-suspend-r4 kernel that I used with Gentoo does seem to cause some issues with the Realtek driver. I had to tweak some settings in the header files for the Realtek driver to get it to compile however I'm noticing that it is now causing OOPSes in 'dmesg' when I try to load the driver. These OOPsies are causing issues since the NIC refuses to come up during these problems.
Fixing it
Wait.. as usual for some patches to roll in. At the current moment it seems that the fixes for the JMicron are in 2.6.18-mm or something branch and will hopefully make it into the 2.6.18 release. Glad to know in 2006, device driver issues still plague Linux. Perhaps one day, everyone will just submit patches into the mainline kernel and will eventually be synced up so people will just have 'working' drivers for their desktops but perhaps that's being too wishful Sun, 13 Aug 2006
How to make a DSP hardware project be on time
Here is an article describing a fairy tale (sadly) idea of how one would run a project for creating a product from DSP chips. I'll have to say it makes a lot of sense but what would be more interesting is some hard data on anyone that has actually managed to apply this idea:
This is OK for a simple algorithm, but what if it is even moderately complex?
Then you are dealing with (i) a very complex algorithm and (ii) very complex
assembler all at the same time. This is usually too much for mere mortals (and
even minor deities). The result is long, painful development, failed projects,
late nights, angry spouses, and lots of pizza (well its not all bad I guess).
The trick is to divide and conquer. Make sure we are only hitting one tough
problem at any given time.
One approach I have found very useful is bit
exact porting to assembler.
There are two important steps:
(i) The algorithm is OK and
(ii) The fixed point port works OK.
In particular (ii) is very tough, so its nice to handle this in a relatively
benign environment like a PC or
workstation.
Read it yourself
Kudos to Joel's Reddit page for this one.
The state of software engineering
If you're curious on one person's viewpoint on software engineering as a discipline (academic and professional) this is a very nice summary / rant on it. One minor niggling point is that I think the author could spend a little time fixing small grammatical mistakes. Since the mistakes make a serious criticism like this look a little childish. But only on the surface. Here's a choice snippet:
First, let's touch on requirements gathering. No where in the book did Pressman illustrate how most engineers actually get their requirements. He presented some idealized scenarios, and correctly illustrated the benefits of use cases. But in the real world, most requirements come from emails and rough wireframes. Assuming that we can start by writing the specification is folly. What we should be studying is how to turn a screen shot created in Photoshop in to a real specification. Engineers need to learn how to annotate a screen shot with input validation rules. They also need to error messages and edge cases. And they need to do it with the understanding that their stakeholders don't know or care about such matters unless you bring it up to them first.
Read it yourself
Kudos to Joel's Reddit page for this one
Bruce Eckel takes another look at Ruby
Bruce Eckel of Thinking in Java fame takes a second look at Ruby. This time he has far less harsh words than his previous look at it (Note that the original post that Eckel made on Ruby seems to be MIA on his website. I'm a little miffed at that...)
His second time around he seems a little more impressed but I have to agree with some readers who pointed out that if he bothered to read the Pickaxe book he would have found a good portion of these features in Ruby. Although for a counterpoint, I've tried reading some comparisons of Python and Ruby and most of them end up in a 'mine is bigger' argument rather than try to really see what is different. Mr. Eckel has spent some time summarizing the differences. The reason it's easier to believe him is because he is staunch follower of Python.
Thu, 10 Aug 2006
Title says it all...
Thanks to Steve for an amusing icon
Tue, 08 Aug 2006
How to downgrade KDE in Ubuntu
In the event you bork your Ubuntu setup trying to move to a bleeding edge version of KDE and want to go back... Read here
Sun, 30 Jul 2006
Some more comments on Mac->Linux switchers
Reasons? The Mac community is too rabid.