Rotating Header Image

Copy from old blog

As promised, here are the blog entries from my old blog.


Ah the wonders of technology, at least they help to dissuade the ire of delayed travellers. On a briefly personal note, I’m currently stuck at an airport, my flight delayed by 5 hours. Fortunately, however, VISA and internet are there to the rescue. After checking if I could find any wireless networks I was rewarded with a dynamic IP (this is a small airport in the middle of nowhere, I hadn’t expected it). After filling in the passport to all doors, my VISA-card number, I am in and able to let the time pass by a bit more pleasantly.

I’m still undecided about the claim I made in my previous post regarding the fact that posting blogs that are more personal is uninformational. In a way, whatever one writes is conditioned by their personality, state of mind and body of knowledge. And some posts, specially those that are more philosophical in nature, will most likely rise from a more personal tale instead of rising from nothingness in an objective statement of their own. Perhaps I should rephrase my position in saying that it is of less (rephrased from ‘no’) interest to read about personal feelings and the likes.

Anyways, time like this is great for writing papers (which I should be doing) as all the other distractions are removed. So with that thought I’m finishing this entry to get some work done.


I really like my new layout. Thanks also Klaas for his insightful CSS tips which cleaned it up a bit. He was also telling me to completely shift away from table’s but I’m too lazy for that and will stick with this for the moment as it works for the time being. I did notice something. My blog page was showing differently, the font was much bigger. In the end it turned out the be the first xhtml remark at the top of the page before the html tag begins. That, I must say, I found rather appalling.

On a briefly more personal note, guess who I heard from today \o/. It’s always nice to hear from her, and if you’re reading, hi sweetie!

Anyways, returning back to more technical aspects, as this blog is after all for that. (I might go into detail in some other blog perhaps why I don’t think personal blogs are that interesting, though to some this might be obvious already). Currently I’m using blosxom for blogging purposes. It can do online as well as offline generation of blogs. It’s rather simple, just requiring some html files and some settings. I generate them on my harddrive and then just rsync my entire home-page darcs repository to the university server. The reason for this is that I am unable to run cgi on the university server, not that this is bad, I find the current solution simple, sleek an elegant. However there is one issue with this, it is impossible to have comments in reply to my blog.

There are several different solutions to this, and if anyone has any other solutions by all means feel free to share them :) .

  • I ask people to email me and select out the most interesting remarks and put them in separate blog entries as trackbacks (or possibly even hack the blosxom system so that it can be linked in more cleanly)
    Pros: Requires no technical challenges
    Cons: Requires a lot of work, will most likely limit my response rate.
  • A second option would be to use the blog I started on blogspot. However I’m rather enjoying the current setup where everything integrates so neatly.
    Pros: Easy to do
    Cons: Spreads out the content of my site
  • Within a similar line, I could move this entire site to my personal homepage. Truth be told it doesn’t appeal to me so much.
    Pros: Would probably work after getting a proper cgi to fit nicely in my template
    Cons: I am not happy with the service provider and they only allow ftp access (which I can’t get from work). The current system or rsyncing is rather spiffy and clean.

Anyways, as always suggestions are welcome, I would prefer if people used my gmail email for this. I’ll most likely build a filter so people can easily comment to my blog and I’ll then categorize it separately so I can easily take stuff and put online.

Using darcs

I was trying to get back a really old version of a darcs repository of a tool I am working on. The reason I wanted to rollback so much is that because the really old version has a GUI although it has much more limited capabilities. Anyways I found that unpull and unrecord didn’t work correctly for some reason. I was informed that it was unable to pullback for some reason. After talking on #darcs and having them tell me to run ‘darcs check’ I noticed that the repository was correct. Therefore they suggested that I send a bugs notification. I think it has something to do with a file having moved around. I’ve done that quite a bit in this repository as I reorganized the different compiler parts and override filenames with new files where old files were residing. For instance:

darcs add AST
darcs mv AST.hs AST/AST.hs
touch AST.hs
darcs add AST.hs
darcs record

Anyways I mentioned my darcs-apply tool that copies a forest of darcs to another forest of darcs, prompting in the case a repository does not exist in the destination forest. They seemed interested but suggested I use a different name. I will most likely move the code from my personal wiki at to here, because I’m not too happy with the clutter of the wiki. They did however suggest that I change the name of the little tool. Maybe Multidarcs? Comments are welcome at my gmail email :)

Lastly I suggested that they should enable pushing and pulling from remote scp darcs repositories without having darcs remotely installed. They suggested I just use rsync instead, as it would be more efficient than doing a pull of the remote repo, applying the patches locally, and then resending the repository. They said that if someone coded it they’d probably put it in darcs, but it seems to be a low priority. I guess I’m stuck using rsync to uni for the moment then, oh well.

Getting started

Well finally, thanks to dons I have my website up. I still need to modify it a bit as I will most likely take another background image. After all having a complete clone is hardly a good idea.

Be Sociable, Share!

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>