No so a buzz, just magic

Actually i wish i could give it a try but let me play with AJAX stuff is not in the future plans of the guy who own it.

Following Tapestry user's list i got noticed by a new framework from Google which by a first glance doesn't sounds too attractive to me since they pretend to compile my code and produce HTML+JS (which builds AJAX) from it.

I whish i write the code no you. They said: "You write your front end in the Java programming language, and the GWT compiler converts your Java classes to browser-compliant JavaScript and HTML."

Well i've to see how it fits... more to come.

I wish i meet Murphy

Well, i don't know if i owe him a big demijohn of good wine for his own thoughts or a big punch on the nose.
I hate when it comes to me but that's sure Murphy got good reasons to speech that way.

Release time it's when you count on things your sure they will not work.

Like the first time

Yes... a deep breath and a deep jump in to the game

Thank you guys!

Web side of the moon

I've stated a lot of times that letting the HTML side of the work done by others not me was simply a dream.
Well this has come in front of me lately, I've developed an almost trivial web app which is based on CSS without touching any HTML word, that was (is since this is only 1.0 and will follow 1.x releases) an amazing way of work.

Kudos goes to my co-workers web guys which were able to work with part of HTML, not only the whole doc,but also (most of the) to the neat way Tapestry and HiveMind, overall, makes you do your own work.

I hope to reproduce this way of develop, yeah i know the gui was really trivial but it was an effectively funny job and things start to get funnier now on web side of the moon.

The commit bit

After submitting my latest patch to Mule i got invited by Trevis Carlson to be a committer to the project, he is the committer who imported my previous patches so I felt really good when received that invite so i accepted.
I owe Trevis a big thanks for all this.

Now i feel i can commit some useful to the project, just need to take time to get in touch with Mule's dev team practice, the next couple of months will be ver interesting.

HiveMind and Mule updated

I've updated the Mule support to HiveMind as a container for its managed components to be able to use the HiveMind service-id as a reference.
This is far more concise the the full class name and overall lets you abuse of it and have (under different service-id) different implementation of the same interface.

After stressing this in my current project i would like to consider having Mule to embrace a Registry supplied by an external entity.
Yep, i can already hear someone thinking to Tapestry...

That's bile!

Yes, that's what i call bile, and there's more and more frightening and frustrating, it's all nothing but true.

Let's say, no bile for nothing.

The mule in the hive

I'm glad to have had Mule to support HiveMind as a container context.

Now I'm alble to use all my services easly, just have to drop my jar packaged service in the mule classpath and it will be published in the registry for me to use it at my will. That's neat.

Funny is that I've never really understood what the name HiveMind was referred to but after have built some services it seems to me to suite perfectly with what the framework actually does, my services act as bees in a hive each coordinated with every other bees in the community.

I'm really pleased to work with this framework.

Analytics... flooded

Well, that's a very smooth service and as result they have been flooded by an amount of request they've been not able to fullfill.
That's what happen when you have cool idea and are able to put them to work.
Now they stopped the service for new users till they add more power, i was lucky enough to have signed up an account early.

Get rid of HTML stuff...

It's time to do some patchworks with Tapestry. Prototyping is finished the time to get dirty fingers is arrived and i get my own fingernails to scratch it.

I've always think to have my web designers produce HTML and sign a contract with them on the ids components should have to let me manage it, but I've found some practice (best one?) from Howard examples show a way to work which makes this impossible.
Let take the Border components from the template guide which everyone should use, i agree that's a good practice from a programmer perspective but my web designers would really like to see their own templates on every browser variant so they've to have the full template not just the snippet one, this cause me work when they produce new HTML output or they slightly modify one.

I'm starting to think I'll never get rid of this...

To embed the creepin thoughts

Well i recently have found time to conduct a very interesting survey on embedded system and solutions.

I've focused my time on OpenBSD and FreeBSD systems as they are the most familiar for me and as they both provide the necessary tools to build (easy) custom systems for embedded environment.

I hope to be able to share my experience, even if not completely.

By by Mac

My new and flashy iMac G5 leave me alone last Saturday morning.

Now it don't even boot, it just stop with a blank black screen, now i've to try some vodo from the Apple Care Service which point me to and (not so) hidden vodo-formula to make the mac come back to life.

I'll tell you.

Voodo is always voodo!

What's up?

Apple get Intel-ized ? Really ?

Did you tell them i got an iMac 15 days ago ?

Getting iMac-zed

Well after more then 4 years i got my OS X.
Ordered on monday arrived on friday, cool.
Opened the box plugged two (yeah i said two) cables and switched it on, it has found all my tech stuff (even the mobile phone), cool.
Never had the experience of using a Mac so i've to get used to but the feelings are cool.
The iMac is decidedly cool, and the OS even cooler.

I even wonder why it takes me so long to get one.

Busy time

I've been busy working a completely different project, just a no java project.
I've been working on making an embedded solution based on OpenBSD.

This is a completely different kind of project, just not a typical programming job, I've just had to be sure a wide variety of stuff work together, the big part of the work has been done by others.

I've said "just" but that's actually a not easy work and definitely not a java project which was my kind of typical project for the last 4 years,this is also the reason it takes me some time to get rid of some strange behaviour but now the work is almost done and i can come back to what i was doing before, port a java app from BarracudaMVC to Tapestry/HiveMind.

For quite a while i will be working on both projects, since working on a project to produce embedded system has made me more and more curious on this kind of job, i must say I've liked that job.

To AJAX or not to AJAX

Creeping around a bit i think it's definitely time to ajax.
I'm following tapestry development which focus on implementing portlet support as the main new feature, well i can say the ajax examples that are popped up in the near past on the tapestry-user@ are very good and very promising, they show how to handle ajax in a pure tap-alike path.

Having a progress bar indicating how far your jobs is going without forcing you to reload the entire page is one of the most interesting thing i could ever wished to have, i even think this is better then portlet support (at least in my web apps world), and it come almost with no cost.

The result is pretty neat and show, once again, how tapestry is well designed and constructed.
More to come...

80% of the problems

I'm almost sure.
80% of problems in custom software development are derived by interface problems. I mean interface by humans.
Don't trust or (even worst) don't understand how custom users wish their program to behave is, by large scale, the first cause of software failure.

So before to look for good developments policy, strategy, tools and programmers just be sure to have the right interface with your customers.

Looking for lighter desktop

I've always used GNOME as my base desktop system but recently i was told to give XFCE a try cause it should be more lightweight and, by consequence, faster.
So i've switched to XFCE 4.2.1, well fonts are smaller and they looks cool but beside that i cannot notice that amount of speed i was told, i'll look further.

Exception day...

Today seems to be like the Exception day for some of the well known java site around.
Forrest gives you welcome with a 500 Internal Server Error while TSS expose you some gateway problem with:

An application exception has occured.
java.rmi.RemoteException: EJB Exception: ; nested exception is: kodo.util.DataStoreException: ERROR: canceling query due to user request {prepstmnt 3446 SELECT t0.threadpk, t0.jdoversion, t0.creationdate, t1.forumpk, t1.jdoversion, t1.rssfeed, t2.categorypk, t2.jdoversion, t2.imagename, t2.summary, t2.title, t1.forumhref, t1.imagename, t1.lastupdated, t1.messagecount, t1.sitepk, t1.summary, t1.threadcount, t1.title, t0.hot, t0.imagename, t0.lastupdated, t0.messagecount, t0.subject, t0.summary, t3.messagepk, t3.jdoversion, t3.body, t3.creationdate, t3.noiselevel, t3.replycount, t4.messagepk, t4.jdoversion, t4.body, t4.creationdate, t4.noiselevel, t4.replycount, t4.parentmessagepk, t4.subject, t4.summary, t4.threadpk, t4.userpk, t3.subject, t3.summary, t5.threadpk, t5.jdoversion, t5.creationdate, t5.forumpk, t5.hot, t5.imagename, t5.lastupdated, t5.messagecount, t5.subject, t5.summary, t5.messagepk, t6.userpk, t6.jdoversion, t6.address1, t6.address2, t6.admin, t6.budgetcode, t6.city, t6.company, t6.country, t6.created, t6.developercode, t6.email, t6.employeecode, t6.fax, t6.firstname, t6.industrycode, t6.interest35, t6.interest21, t6.interest30, t6.interest17, t6.interest2, t6.interest13, t6.interest37, t6.interest34, t6.interest32, t6.interest33, t6.interest22, t6.interest11, t6.interest10, t6.interest9, t6.interest24, t6.interest19, t6.interest12, t6.interest36, t6.interest25, t6.interest16, t6.interest4, t6.interest8, t6.interest26, t6.interest7, t6.interest20, t6.interest31, t6.interest6, t6.interest15, t6.interest5, t6.interest23, t6.interest3, t6.interest28, t6.interest14, t6.interest29, t6.interest18, t6.interest1, t6.interest27, t6.last_login, t6.last_modified, t6.lastname, t6.password, t6.phone, t6.province, t6.scopecode, t6.state, t6.title, t6.titlecode, t6.username, t6.verified, t6.zip FROM public.threads t0 INNER JOIN public.forums t1 ON t0.forumpk = t1.forumpk LEFT OUTER JOIN public.categories t2 ON t1.categorypk = t2.categorypk LEFT OUTER JOIN public.messages t3 ON t0.messagepk = t3.messagepk LEFT OUTER JOIN public.messages t4 ON t3.parentmessagepk = t4.messagepk LEFT OUTER JOIN public.threads t5 ON t3.threadpk = t5.threadpk LEFT OUTER JOIN public.users t6 ON t3.userpk = t6.userpk WHERE (t1.forumpk = ? AND t1.sitepk = ?) ORDER BY t0.creationdate DESC LIMIT ? } [code=0, state=57014]

Even if TSS is running Tapestry...

Will java go for *BSD

I really cannot get why we cannot get a good binary ready version of java on BSD.
I really cannot get which the problems are, it's really a matter of technical issues ?

I've followed the road of java on FreeBSD since ealry 2002 and now almost three years are gone but there's no light in the tunnel.

Are actually the guys at apache eating their own food ?