All posts by viz

New Applications on Top of the Web – LumberJaczk is coming.

I’m really looking forward to announce my project I’ve been running since the last summer. And it’s about to happen soon, really soon :). Its mission has crystallized to be: New LumberJaczk Project

An open-source technology that enables better ways to get and use information on the Web. Let people create light-weight on-top-of-Web applications of their own and share them with others.

Hacking out such use-cases like “best flights search”, “personal investments monitor” will be fun to create. At least for me it definitely is :) All based on Firefox browser and it’s Mozilla guts. BTW, the project name: LumberJaczk.

So where are we now?
The system has already been functional and ready to use in alpha version for some time. What keeps me busy right now, more than I expected, is preparation of all sorts of presentational stuff to communicate the project to public clearly and concisely. Presentations, screen-casts, docs are ready and the web page is in progress. My estimations are the first come-out of the Web is to happen the next Mon, May 21 at So check back, subscribe or whatever… I’m looking forward to hearing some feedback once it’s out. After that another big thing is to make a code due-diligence so the code release can follow soon.

Mozilla DOM W3C Connector announced

Mozilla DOM to W3C ConnectorJust an information: Peter Szinek, a friend of mine is about to release the Mozilla DOM W3C Connector he implemented. It allows you to access the Mozilla DOM as W3C DOM from within Java when embedding Mozilla. This is fundamental if you want to use standard XML tools and libraries to operate directly on Mozilla DOM, e.g. XML, XSLT, XPath, XQuery. Read more in his original  post.

We’ve been using it for few months and it proved to be stable. The bad guy seems to be sometimes JavaXPCOM ;)

A Year with SCRUM

It’s been a year we implanted Scrum (or something Scrum-like) as our development management methodology in the company that employs me atm. Today we said goodbye to the 12th sprint/month, which was a great chance to make a “1st birthday of Scrummy boy” party as well. Besides the regular Sprint-end presentation I presented also a look-back yearly review. Has it worked ? What do we do differently ? What have we learned ?

Year ago

We were in a situation that reached some critical-mass. Maintenance of actual products – unfortunately quite complex, running critical new development, prof. services requests hitting us daily, ever changing priorities, recursive interruptions, ill communication, organizational changes. All this with limited resources. Stress and frustration for every one. Gannts and quarterly planning surprisingly didn’t help ;)) We thought we did extreme programming. We did extreme… just in other aspect :) A small dev-team uprising was unavoidable.

…blah..blah…and so we decided to go with Scrum as project management tool. Backlog, ScrumMasta, Sprints, Daily scrum… Because anything was better than the status-quo the installment was smooth and accepted. Personally, I don’t like very much the way Scrum is promoted. And for the book: Ken Schwaber: Agile Project Management with Scrum… ok, but no big thing. Importantly, what is very sound and pragmatic are the the Scrum concepts. Of course, as we are somewhat creative people, we could have never been orthodox followers. Moreover it’s just about project management so you have to complement and stir it with other concepts (An Agile Roadmap)to get the most tasty and nourishing soup for the complete development process.

Year after

We are team of 6 devels, quite cross-functional. Very flat structure. New development vs. professional services,support, and maintenance goes 1:1. Besides the basic principles, we came up with some Scrum patches and local customs:

  • ScrumMaster is one of the devel team, as his part-time job. This eats max 20h of his time, which is sufficient for the duties.
  • Rotating Scrum Master -1-3 times each; to avoid stereotype.
  • PPTS – The tool. Web based, GPL, SCRUM, extreme programming. Very useful.
  • Scrummaries – each member writes a mail with a brief description of his achievements, status, open points at the sprint-end. It’s helpful later on to have such info persistent.
  • Half-time round-trips – SM and PO has a short talk to each devel get the big picture view on the progress. Some steering is usually done here.
  • Emergency time slots – we experience cca 15% of unpredictable situations, so-called emergencies. Therefore, we add a constant-time slots to each sprint.
  • DailyScrum mailinglist – as the team cannot be daily together, the missing person ports the “daily answers” to the mailing list.
  • Bad task” – unpredictable, NP-hard, R&D, we fix time per sprint.

After all, I have to say this approach worked for us quite well. The major improvements came in process transparency, much better planing, much much more relaxed development and thus also efficiency. And not so “agile” as before :)

So can I recommend this? Look at your problems, your team, the project and company culture. And make your judgment yourself.

Text Manipulation in Gimp

I’m running Ubuntu Breezy. The latest Gimp available – 2.2.x still suffers from insufficient text manipulation functionality e.g. letter spacing, transfomations. It seems some improvements will be available in 2.4. Till then a Freetype plugin can help a bit.
The page offers just a source so you need to compile and install it:

  1. Get the source code
  2. apt-get install libgimp2.0-dev
  3. Untar source/configure/make/make install
  4. Start gimp, you'll find it in Filters>Text>Freetype
  5. Add path (configure) of each font you want to use separately. Adding just /usr/share/fonts is not enough.

How is the XPather doing?

Recently, a friend of mine asked me how is my XPather doing (for those who do not know it, XPather is a Firefox browser extension that generates XPath of DOM elements either directly from the browser or while inspecting the DOM in DOMInspector. Further, it evaluates XPaths, works with relative XPaths, extend filtration by regexps and allows for simple content extraction.)

So the friend picked my brain for some stats like downloads, user-base, support, new plans … To be honest, he surprised me. I was pretty busy for the last few months so ehm… I didn’t really know somehow. Right that evening I took a look into it and found out the following:
The XPather is now in its 7th month since the first release hosted on and on its project home The number of downloads hits 10500 this week. It would be also very nice to see how many active users are out there. The first approximation here could be to have an average count of daily update requests. Although many users has automatic updating turned off, while others are just “quasi-users” that just installed the extension (like 10000 other extensions they tried out) but have never used it. The only numbers I can get is from the XPather homepage: 35% of the downloaded installations are still active. Would be nice to get such statistics from As for user feedback and comments, I have to admit I expected a bit more :) Anyway. It was a nice finding, after all, the people are still pretty interested in data extraction, screen-scraping, greasemonkey scripting etc. Personally, I still use the tool for quick hacking of web pages or when I work with XML.

And for the plans ? Well, I consider the project to be in maintenance mode as I focus completely to a new project called LumberJaczk. But still, if you people have some features you’d like to see in the XPather, or find some bugs, let me know. I’m here to fix it.