Comments for Dariusz Cieślak's Blog on Software

Comment on “Usability” na przykładzie bankomatu by prad

prad — Thu, 19 Aug 2010 09:59:52 +0000

Gdzieś pisali też chyba o tej reklamie że cudzoziemców to już na amen zmyla – skąd oni biedni mają wiedzieć co to TAK/NIE? Kasę chcieli wziąć tylko.

Comment on How To Debug JNDI Lookup Problems by Fokko Groenenboom

Fokko Groenenboom — Fri, 11 Jun 2010 06:32:04 +0000

${WAS_INSTALL_ROOT}/bin/dumpNameSpace.sh should do the trick also

Comment on Twitter is over capacity / 502 Bad Gateway by flaker.pl nie działa / 504 Gateway Time-out « Dariusz Cieślak's Blog on Software

Wed, 09 Jun 2010 05:11:39 +0000

[...] widać serwis boryka się z problemem wydajności podobnie jak Twitter. Czyżby znowu zima zaskoczyła “drogowców” ? Comments [...]

Comment on GoldenLine nie działa by m

Fri, 04 Jun 2010 21:41:23 +0000

ja rowniez nie moge otworzyc stronki http://www.goldenline.pl nie wiem czy to tylko moja wina, czy Ty rowniez masz nadal problem??
dzieki za wszelakie info

Comment on MySpace vs Facebook vs Twitter uptime comparision by dariusz.cieslak

dariusz.cieslak — Thu, 20 May 2010 21:10:50 +0000

I appreciate their (Twitter team) efforts and their optimisation task size (enormous hit count per second) but it seems the variance is high from few months not days. Looks like Ruby-based architecture optimisation wasn’t a simple task.

Comment on MySpace vs Facebook vs Twitter uptime comparision by Michał

Michał — Thu, 20 May 2010 21:00:27 +0000

Yeah, saw quite a few fail-whales in the last few days. API isn’t working too well either. Maybe Twitter should drop Ruby and use a proper technology like PHP

Comment on Is Static Typing the Root of All Evil? by Michał

Michał — Mon, 17 May 2010 19:24:15 +0000

While learning programming I once was moving from PHP, which has pretty much no typing, to Java, which has full typing. Back then that move was all but easy and I kept complaining on why Java has to make it so hard to manipulate data of various types. I did get through that though and now I find it hard to work with typeless languages again.

It’s simply that Java offers many constraints, thus allowing programmers to apply plenty of checks for whether data passed on is correct. This in turn protects us from hurting ourselves accidentally and thus we can code with more certainty that we won’t break anything that elementary.

Working with ColdFusion right now I constantly need to use functions like IsArray, IsBoolean etc. Checking for whether my data is what I expect it to be.

Comment on Static Verification for J2EE-Based Software by Is Static Typing the Root of All Evil? « Dariusz Cieślak's Blog on Software

Is Static Typing the Root of All Evil? « Dariusz Cieślak's Blog on Software — Sun, 16 May 2010 12:55:04 +0000

[...] project-specific static analysers [...]

Comment on ORA-00600: internal error code by calel0722

calel0722 — Mon, 03 May 2010 23:27:38 +0000

I have the same problem, try recreate the indexes on the relevant table
If you have tha solution, tell me.
Good luck!!!

Comment on Websphere Extreme Scale + Hibernate = ? by Talip Ozturk

Talip Ozturk — Sun, 02 May 2010 21:54:06 +0000

Also check out Hazelcast!

Hazelcast (http://www.hazelcast.com) is an open source, transactional, distributed caching solution for Java. It also has hibernate second level cache plugin and is released under Apache license.

Hazelcast is actually a little more than a distributed cache; it is distributed implementation of queue, topic, map, multimap, lock, executor service for Java.

Comment on “Secure Connection Failed” on https://forum.hibernate.org by dariusz.cieslak

dariusz.cieslak — Fri, 30 Apr 2010 19:44:10 +0000

I trust no DNS servers I don’t administrate. Any local network sysadmin is able to “poison” local DNS proxy and redirect you to mirrored bank portal. Without server certification it would be very easy to do.

Comment on Websphere Extreme Scale + Hibernate = ? by Michał

Michał — Fri, 30 Apr 2010 19:26:02 +0000

Yeah, very good conclusions, and ones that deserve and extended article I’d say. My last company has a policy of choosing closed and paid-for software, so we’re using Perforce (instead of say SVN or something more modern like Mercurial) and DevTrack (instead of something more reasonable like Mantis). We’re not satisfied with any of these, but hey – we have support!

Comment on “Secure Connection Failed” on https://forum.hibernate.org by Michał

Michał — Fri, 30 Apr 2010 19:14:30 +0000

These warnings are funny. Has anyone *ever* seen one, which was ‘real’? I mean one which really appeared on a malicious website or during a man-in-the-middle attack?

Comment on Why svn:mime-type does matter? by Why svn:mime-type does matter? « Dariusz Cieślak's Blog on Software « svn

Why svn:mime-type does matter? « Dariusz Cieślak's Blog on Software « svn — Fri, 23 Apr 2010 15:53:34 +0000

[...] Więcej: Why svn:mime-type does matter? « Dariusz Cieślak's Blog on Software [...]

Comment on Rezerwacje zmiennej długości w med.aplikacja.info by dariusz.cieslak

dariusz.cieslak — Fri, 09 Apr 2010 10:12:48 +0000

Proszę bardzo:

Reservation += [visitTimeMinutes, service -> Service]
Service [name, visitTimeMinutes].

Redundantny Reservation.visitTimeMinutes wbrew pozorom przydatny.

Comment on Rezerwacje zmiennej długości w med.aplikacja.info by Michał Paluchowski

Michał Paluchowski — Fri, 09 Apr 2010 10:04:08 +0000

Ciekaw jestem jak to od stronty modelu wygląda.

Comment on Is “commit freeze” really required for software development? by PR

PR — Tue, 23 Feb 2010 22:03:29 +0000

Yet another reason for freezing commits – lazy&remote SCM engineer. However, freezes _might_ have some psychological impact. Both good and bad.

Comment on Subversion Best Practices by PR

PR — Tue, 23 Feb 2010 22:01:11 +0000

For pre-1.5 SVN era you could use svnmerge – http://kenkinder.com/svnmerge/
There are some differences in comparison to merge tracking in Subversion 1.5 (both pros and cons). But it does its job. In the past it did even in one of the project you participate yourself – SCNR

Comment on Peer code reviews with Subversion by Gregg Sporar

Gregg Sporar — Mon, 08 Feb 2010 00:16:47 +0000

Agreed: Peer review plays very well with other agile tools. More thoughts on that topic here: http://blog.smartbear.com/the_smartbear_blog/2009/11/peer-code-review-an-agile-process.html

Comment on Peer code reviews with Subversion by Jesse Gibbs

Jesse Gibbs — Sun, 07 Feb 2010 15:21:59 +0000

Good article, and good recommendations. Peer code revew tools such as Atlassian Crucible, Smart Bear Code Collaborator, ReviewBoard, etc. facilitate the exact process you mentioned above, and they keep you from having to 'pollute' your code base with a lot of comments. Crucible supports threading of comments (similar to a forum discussion) and integrates nicely with FishEye for performin diffs and other visualisations. Plus you can create and manage defects as JIRA bugs with only a few clicks, to ensure bugs found during code review are actually fixed.

Comment on Pytanie za 10 punktów: kiedy Onet.pl “śpi”? by o.

o. — Wed, 27 Jan 2010 07:29:21 +0000

bzdura.

Comment on Recenzja VPS w vpsinn.pl / statnet.pl by admin

admin — Sun, 03 Jan 2010 20:36:17 +0000

Rzeczywiście, pomyłka na korzyść klienta to sygnał ostrzegawczy – w firmie może być bałagan (zobaczymy jak będzie układała się współpraca).

Z S3 korzystałem (wynajmowany storage na potrzeby jednego z projektów), z EC2 nie miałem okazji (zniechęca mnie “ulotny” charakter maszyny wirtualnej – baza danych musi już być poza VPS-em na oddzielnej usłudze).

Comment on Recenzja VPS w vpsinn.pl / statnet.pl by Michał

Michał — Sun, 03 Jan 2010 16:57:22 +0000

Dziwna to trochę sytuacja, kiedy oferują określoną ilość pamięci i rdzeni procesora, a udostępniają więcej, nie pisząc o tym oficjalnie ani słowa. Rodzi to podejrzenia, że zaszły jakieś pomyłki – tym razem na korzyść klienta, ale za chwile być może podobne zajdą na niekorzyść? W każdym razie skoro już z nimi jesteś, to czas pokaże zapewne na ile się sprawdzają.

A tymczasem, co z Amazon EC3? Rozważałeś pewnie ich wykorzystanie, ale czy próbowałeś?

Comment on Aptitude vs apt-get memory usage on Debian by Jannich

Jannich — Wed, 30 Dec 2009 13:31:57 +0000

Interesting.. I think I’ll stick to apt-get on the vps’s then.

Comment on Linode VPS, welcome to Europe by Killer Bee

Killer Bee — Sun, 27 Dec 2009 04:04:17 +0000

Linode is cheap vps and I read some article about it and they said that Linode is the best.

Comment on Subversion Best Practices by admin

admin — Mon, 14 Dec 2009 08:01:24 +0000

Thanks. Indeed subversion 1.5 handles merge information using properties.

Comment on Subversion Best Practices by Michał

Michał — Sun, 13 Dec 2009 10:56:35 +0000

Subversion doesn’t lose commit information on merge. At least not since version 1.5 was released, including merge tracking. As of that version you can easily see all the commits and authors behind merges.

Comment on Ramhost.us VPS review by RAM Host

RAM Host — Wed, 14 Oct 2009 22:42:46 +0000

Thank you for the comprehensive review.

Your signup delay was most likely caused by ordering late at night (we don’t wake up staff for new signups – during normal hours we provision service in ~30 minutes).

I hope you continue to enjoy your vps with us. If you have any suggestions or comments please feel free to let us know.

And once again, thank you for the review.

- Admin, RAM Host

Comment on Apache to Lighttpd migration by admin

admin — Wed, 16 Sep 2009 19:16:08 +0000

Unfortunately production installation with Lighttpd on a VPS (a2b2 disaster few months ago) was moved to shared account, so currently no benchmarks are possible. I hope to install LH on my next VPS and will share some measures.

Comment on Trac/Agilo software helps track bugs by admin

admin — Wed, 16 Sep 2009 19:12:29 +0000

Yup, my Agilo experience was narrowed to Track-related functionality. Burn-down chart was managed independently by external tools and track reports, so didn’t see added value of Agilo. Maybe in next project …