Weblog Without Honor or Humanity

Looks like someone is reading this blog after all. Yesterday I wrote about new collaborative text editor for Windows called multi-editoro and a reader pointed me to DocSynch.

DocSynch is an interesting approach to building collaborative editor: it only implements a protocol layer based on IRC and plugs into existing editors. Currently they have a plugin for jEdit and are planning additional plugins for Eclipse, vim and Visual Studio.

Another day, another text editor to check out.

SubEthaEdit, a collaborative text editor for Mac OS, received a lot of buzz and for a good reason. It does basic text editing well, is a well designed application and adds a killer feature not available in other editors: collaborative editing.

Unfortunately for me it only works under Mac OS while I mostly use Windows so I was looking for similar Windows app. multi-editoro is such an app. I only tried it for a while for non-collaborative editing so can’t yet tell how it works for its indendent purpose but I do have a few comments on the app.

The app is much less polished that SubEthaEdit. Even though it’s a Windows app, for the UI it seems to be using some sort of curses-based interface. This is not acceptable for a Windows app. It’s ok for toy apps but not ok for end-user apps intendent for average users. I have no doubts that the reason for that is that it’s much harder to write UI in native Windows and that Unix version is also possible but users don’t care.

The main page on a website is a big flash animation. Seriously, when will people learn? You would think that since books like Philip and Alex’s Guide to Web Publishing or resources like Alertbox are available for free on the web, people doing websites would read them and stopped using Flash gratuitiously since it makes website less usable in many ways than a simple html page, not to mention the maintenance problems for webmasters when they have to change anything on the website.

A better name wouldn’t hurt. There’s already at least one multi-editor out there and one letter difference in a name isn’t quite enough.

An installation file should also be provided - currently it’s just a zip file.

By default, the text I write should not be highlighted.

So the final verdict is that multi-editoro is an interesting proof of concept that is worth keeping an eye on but it’s far from being a usable, end-user application.

I cannot be 100% sure but it seems that the author wrote his own text editor engine which to me is a big waste of time. Writing a high-quality, bug free text editor and display engine with a lot of features takes years. Collaborative editing part is comparatively small part of the overall code. I don’t understand why people choose to waste a lot of time re-implementing the code that they can get for free. The best way to get a high-quality collaborative editor would be to just extend SciTe (or some other mature, open-source text editor) and add collaborative features to it.

So is the new version of Treo real? According to this post yes. They guy links to a screenshot of a PDF page showing that new Treo will be available Oct/Nov 04.

This is how competitive pressure works. Undoubtedly as a result of Google’s free 2GB gmail, fastmail.fm just announced that they’re increasing their quotas. This is great news for me - I’m using enhanced account and my mail storage quota will go from 150 MB to 2GB. The good thing about Fastmail is that it allows POP and IMAP access. The good thing about gmail is great web interface. Fastmail’s interface ain’t that bad either but gmail really created something substantially better than any other web e-mail interface that I know. I’ll keep using them both, for slightly different purposes.

First beta of PostgreSQL 8 has been released. 8 is a significant release because it’s the first one to provide native Windows version. Previously only a cygwin port was available and it was simply unusable (and I tried to use - even wrote tuturials about how to do it).

This was long coming. MySQL, PostgreSQL’s main “competition”, had Windows version for a long time. I believe they have more downloads of Windows version than of Unix versions.

This is a very good change. I, for one, chose MySQL over PostgreSQL to implement one of my projects because Windows version was available. Although final deployment was on remote RedHat Linux server, I do my development on Windows. I’m much more productive in that environment. Not being able to use PostgreSQL on Windows, ruled it out as an option for me (even though I believe it’s technically superior to MySQL).

Out of curiosity I’ve installed PostgreSQL 8 on Windows. Installation process was quick and painless - setup is pretty good and and takes care of boring detalis (like creating user account that will run PostgreSQL, installing it as a service etc.)

Too bad that pgAdmin III, a GUI front-end for managing PostgreSQL databses included in the package, is so poor. The version number would have you believe that it’s a mature application but it’s simply outclassed by what I’m used to (Workbench application in upcoming Yukon version of Microsoft SQL Server). pgAdmin III isn’t even very good when considered in a vacuum but when compared to evolved MS SQL Workbench just makes you reflect on how much catching up most open-source application have to do to be of comparable quality with their commercial counterparts.

PalmSource finally released the final 1.0 version of PODS (Palm OS Developer Suite).

I’m not terribly excited - I haven’t even installed it yet. This is, however, the future of Palm OS development. PODS gives everything you need for writing C/C++ code for OS 5 and earlier and is the only tool that allows writing code for OS 6. The editor is based on Eclipse 2.1 and CDT (a plugin for writing C/C++ code) although Palm is planning to move to Eclipse 3 in the future.

It also have C/C++ compilers (uses open-source prc-tools for pre-OS 6 code) and Palm’s in-house ARM compiler for OS 6. As a side note: it’s a strange decision by Palm to develop (or license - don’t know the inside details on who wrote that compiler) an in-house ARM compiler instead of just using gcc (with modifications, if any are necessary). PalmSource is cash-strapped - they clearly can’t afford to hire as many programmers as they should - and it seems to me that taking gcc code and funding necessary modifications (if any) would be much cheaper than writing or licensing the whole compiler.

PODS also includes debuggers and simulators.

PODS is important because it’s the future. Before PODS developers had a choice between CodeWarrior, which includes IDE but is expensive, and prc-tools (free but there’s no IDE and it’s not as good).

Now PalmSource is giving IDE as good (if not better) as CodeWarrior for free so I doubt that Metrowerks is going to try to compete and sell dev tools for Palm OS in the future.

At this point PODS is the only tool that allows writing code for OS 6.

I’m not jumping up and down from joy because at this point there’s no reason to write OS 6 apps - there are no devices that actually ship with OS 6 (despite the fact that it has been officially released at the end of 2003). Even after devices are available it’ll take years for OS 6 to have significant market share and given that it can run pre-OS 6 apps just fine, there’s little reason to hurry.