Planes are like ants trying to find food and auto-organizing themselves (not necessarily with the shortest path). This recall me of the Ant colony optimization for solving TSP problem.

Here are some resources for the ACO optimization for TSP problems: http://www.scribd.com/doc/3689188/Ant-Colony-TSP-Optimization, http://www.ugosweb.com/Documents/jacs.aspx or http://www.idsia.ch/~luca/acs-bio97.pdf.

• Use the search bar:
  this redirect you to your favorite search engine results pages for the keywords entered.
• Use the address bar:
  this also redirecte you to the results pages UNLESS the first result is good enough. If you type in “firefox” or “wiki belgium”, the first result of the search engine is probably what you’re looking for. So why lost time in the results page and click again on the obvious first result?

And for the people willing to know how this works, it is pretty simple. Firefox use this setting “keyword.url” (in the about:config) as the url to query for keyword input. In my configuration, it’s “http://www.google.com/search?ie=UTF-8&oe=UTF-8&sourceid=navclient&gfns=1&q=”. This is the usual google search url at the exception of the param “gfns=1″ which is the “good enough” redirection.

It’s kinda like the “I’m feeling lucky button” but you redirected to the first result only if there is pretty good chance that’s what you’re looking for (above some threshold I guess). And if your query was unclear, for eg “eric rodriguez” (I’ll bet you’ll never find me , then you are redirected to the usual results page to make your choice.

Pretty useful and neat feature, isn’t it? IMHO, it’s the fastest way to seach & surf (so far). Let’s try it and like me, you’ll get addicted to this good enough redirection

Moreover, this is a one shot generation so any “quick and dirty” solution would have satisfied me Here are the possible solutions I found:

• Webservices: there are plenty of webservices to generate webshots on the go. The pricing is quite low and this is definitely a quick way to build your webshots collection. Among others, you can check: webthumb, site thumbshot, Quick Thumbnail, Thumbalizr. You can check google for more…
• Applications: you can also (as I did) rely on simple applications/scripts like CutyCapt, IECapt, Paparazzi, webkit2png or khtml2png. On the first pass, I used CutyCapt, but on second pass I switched to IECapt since a lot of my websites needed flash plugin in order to render properly.
• Build your own server: there are also a couple of articles describing how to build you own webshot server: with python and qtwebkit, with firefox. You can also check thummer(a global webshot server solution) or this article.

For me, CutyCapt and IECapt (with some batch script) did pretty well

1. Focus on Content
   A main characteristic of Twitter is the 140 characters limitation. So be sure to be concise & precise. You can say a lot even with 140 characters! And the best way to grow your followers community is to provide clean and clear ideas.
2. Be polite
   This may seem obvious but there are already some abuse reported. So think twice when you tweet, because your short text will be publicly broadcast and read before you’ve got time to fall back. You’ll find a Twitter Etiquette here or here (if you need guidance;)
3. Do not give away your password
   Last item in the obvious series, do not give your password to third parties. If you don’t know why, just ask yourself if you would like to give your phone and the PIN code to strangers. So now you’re warned, don’t cry for identity theft or messages you didn’t write yourself.
4. Control your tweet speed
   This advice can be harder to follow. When you tweet, you’re not only sending a message to 1 or 2 friends but to a whole community. As each one tends to follow more than 2 friends, It can quickly become annoying when you’ve got a black sheep twittering every move each 5 seconds. If you don’t want to loose followers or friends, keep your tweet speed to a decent level. This way everyone can still communicate instead of hearing some white noise.
5. Use @, RT and #
   There are 3 basics twitter syntax you can rely on:
   
   • @ is the common identifier for people. You’ll use it a lot in directed (yet public) conversation.
   • RT is to use when you forward a message you like and want to share with your own community.
   • The # sign is used when you want to specify an event or subject. By using #, you ease the search or filter use to people only interested in a topic.
6. Choose with caution who you follow
   At first, you will be tempted to add as many friends you can, like you do on Facebook. But the thing is Twitter is no Facebook: someone you follow will “send messages to you”. Imagine you follow 100 persons tweeting 5 times a day, it will be so messy you’ll begin to taste some “I’m getting too much spam” feeling.
7. Get an anti-spam
   You may think I’m silly, but I think security software actors could someday provide a Twitter-AntiSpam. If you think this is a joke, you probably don’t know yet: “viraltweets” or “twitter le truc du jour“. So be cautious when you receive strange tweets, sometimes it is no good to click on every link. And if you get caught by some script retwitting this spam, at least clean you account and delete this unwanted message.
8. Less is More: use shortening tools
   Due to the 140 characters limitation, you may find these shortening tools rather useful:
   • To compress urls in a few bytes: bit.ly, tinyurl.com, snipurl.com (sn.im), is.gd , tr.im,….
   • To share pictures: twitpic.com
   • To compress text: 140it.com, tweetshrink.com, twonvert.com
   • To share source code:  snipt.org

   You can also refer to this article about best practices to short-tweet.

9. Avoid pyramidal traps
   If you still think the more followers you have, the best you are… you will probably get trapped in such pyramidal scheme like TweeterGetter. This website propose some “viral” approach to get followers. But the truth is you give your Twitter account to a stranger that will automatically build a pyramid of accounts. Viral marketing should not be confused with basic chain-letter scheme.
10. Monitor and engage your followers
    Building a large and faithful community, do not hesitate to engage your new followers by thanking and trying to know them. This way you’ll build a win-win relation with people interested in what you have to say or share with.
11. Don’t forget your profile
    Once you’ve got your brand new account, take some time to fill your profile details. It can be bad and at least new followers will have some information about you and your main interests. They may even go to your website, so there is no reason to not share such details. Moreover you can also personalize your Twitter page, so visitors will appreciate a appreciable welcome note.
12. Use adequate tools
    There are a lot of Twitter related tools. Thanks to Xavier from Tweetag, you can print this handful cheat sheet card. This toolbox gather the main Twitter tools you may need. If you need more tools, you can refer to the Twitter apps db.
    My favorite ones are:  StockTwits, TweetDeck, TweetWheel and MicroPlaza (if you’re lucky enough to have a beta invite). There are a lot of others but you’ll find easily other tools lists.
13. Split personal and professional accounts
    It is considered as a good practice to have both a personal and a professional email account. And it’s the same for Twitter! If you really are using a twitter account for business needs, you may prefer to split your personal and professional tweets. I know that social media are all about the person engaging a community, but I’m not sure the “Safe at home at 6h12 in the morning after 2 vodka and 1/2 whisky bottle” tweet will enjoy your professional followers

I hope you enjoyed this best practices and feel free to propose more… See you soon on Twitter @Wavyx.

• Feedback: it’s good to let the customer know that you are (or will be) working on his issue. It gives a warm feeling like “I’m not alone in this mess”. It also gives a more professional look to mail support.
• Centralization: it’s easier to manage issues or requests with a centralized system. After all, support or issues tracking is a MIMO (Multiple Input – Multiple Output). Multiple Customers having problems have to meet the Multiple Developers/Professionals and find the one fitting their needs. Moreover, sometimes  you cannot handle a whole case by yourself from A to Z. Then you will be more than happy anyone else in your corporation can solve and close the case if you’re no more available.
• Knowledge: by using ticketing system, you get a bonus! The information is tracked and persisted so you can find back relevant information when someone already dealt with this problem.

If you agree with this, the only thing left to do is to find a suitable ticketing system. So here comes my short list of 4 ticketing systems:

1. RT for Request Tracker. RT is available since 1996. It’s a wells spread system built in Perl and offering an enterprise-grade ticketing system with a lot of features like identification, prioritization,
   assignment, resolution and notification.
2. OTRS for Open source Ticket Request System. Also built in Perl, OTRS offers the same usual features to manage customer telephone calls and e-mails.
3. osTicket is an open source support ticket system written in PHP.  It offers a web-based solution to easily manage, organize
   and archive all your support requests and responses in the same place.
4. eTicket is another PHP ticketing system, really similar to osTicket (above).

There are of course commercial solutions but I focused on open source alternatives. So you may also have a look to BMC, helpdesk pilot, tenderapp, simple ticket, mojo helpdesk or h2desk.

And if you still need other systems, you can refer to wikipedia, dmoz or slashdot.

For .Net developpers who needs a local databse, you can use SQLite in application developped in C# or VB.net by using one of the following wrappers:

1. System.Data.SQLite is the original SQLite database engine. It is a complete ADO.NET 2.0 provider and can be used as full replacement for the original sqlite3.dll. It has no linker dependency on the .NET runtime so it can be distributed independently of .NET.
2. Finisar.SQLite was the original ADO.NET Data Provider for accessing SQLite-Databases using the .NET-Framework. It is no longer under active development and has been replaced by System.Data.SQLite.
3. SQLite.NET wrapper  is a simple wrapper composed by a couple of classes in C# to allow access to SQLite databases. This is not an ADO.NET provider, but instead it provides a simple set of methods to access and retrieve data in a result set class.

I recommend this 2 quick tutorials to get started with SQLite in a .Net environment: SQLite on .NET – Get up and running in 3 minutes and Using sqlite with .NET.

Last but not least, if you need some visual database manager (GUI), you can try SQLite2008 Manager, SQLite Administrator or SQLiteTool. See also SQLite Trac for a more comprehensive list or this link.

This generally implies a lot of marshalling and interop to wrap unmanaged windows functions into nice managed .Net namespaces. And this is also true for mobile developpers interacting with Windows CE dll, see Interop for .Net Compact Framework.

I just discovered that  a couple of Win32 functionnalities are already bundled in the Microsoft.Win32 namespace of the .Net framework.

Another great resource for COM interop is the P/Invoke wiki !

And for those who are (still) more familiar with the old school windows functions, you can refer to this map to find your way in the .Net environment.

• Griffon is a Grails like application framework for developing desktop applications in Groovy. You can begin with the quick start guide.
  
• Gradle is a build system like Ant, Maven or Ivy but trying to get the best from all. It supports multi-project build and  dependency management. And you still can use you old Ant tasks.
• Easyb is Behavior Driven Development framework (BDD). It uses a specification based Domain Specific Language (DSL).  The main idea is to keep really close to the business needs all along the development process. With this tool you’ll have a readable documentation AND a unit-testing all-in-one. You may start by reading first this tutorial.
• Compass is an open source project built on top of Lucene, to simplify the integration of search capabilities in your java applications .

And of course you can still refer to the Groovy and Grails websites.

1. Microsoft Patterns & Practices provide an Updater Application Block v2.0. An application block is a kind of reusable and extensible module. This specific application block provide a quick and easy mean to add self updating capabilities to .NET-based smart client solutions, like Windows Forms applications. Here you can find the Introduction, the Design overview and an article on how to use this application block.
2. Sharp AutoUpdater is a library providing an auto-update feature for .NET applications. It relies on XML configuration files and handles the detection of a new version, downloads, unzips, and installation of the new files.
3. Microsoft also has a ClickOnce deployment solution that handles updates. It is similar to the  java webstart solution. You can find more information here, in this article or among the wikipedia external links

ClickOnce seems fully integrated in a .Net environment, but as far as I know there is no solution to use a standard webserver (not IIS) to handle updates. I should definitely give it a try and see if IIS is really mandatory or if you can use some workaround.

Do you know other methods? Sometimes deployment tools include auto-update container to bundle an generic updater with your application (like install4J).

Eric Schmidt On Tech Jobs

Yesterday, I randomly discovered a new tool for .Net to help developers with code testing. Microsoft Research has developed Pex, a test plugin which can be integrated to Visual Studio. So far nothing is really new, right ? But, say you’re as lazy as me when it’s about writing unit tests, you should definitely go look the official Pex homepage.
To be more specific, Pex automatically discovers interesting test paths. It finds interesting input-output values for each method and generate a test suite offering a pretty high code coverage. The keyword being “generate”, which means less time on testing and more on… (suit yourself).

At first sight, I must confess this may sounds a bit magical.  So how does it work?  Pex performs a systematic analysis, hunting for boundary conditions, exceptions and assertion failures, which you can debug right away. Bringing all this together, Pex offer a new approach to Parameterized Unit Testing and will hopefully reduce debug time and maintenance costs. This general test technique is also referred as  white box testing.

You will find extra information on this slideshow or on the Pex documentation section. Channel9 offers a “getting started” video as a tutorial on how to use Pex inside Visual Studio 2008. Microsoft Research has also a document about Parameterized Unit Tests.

Also let’s stay cautious: automated or generated tests are not THE solution to avoid code inspection and application stress. As a responsible developer, you should use tools to ease your life and make your products better. But never be too confident when it comes to magic black box software making your job

• Jericho HTML Parser is a java library allowing analysis and manipulation of parts of an HTML document, including server-side tags, while reproducing verbatim any unrecognised or invalid HTML. It also provides high-level HTML form manipulation functions.
• MozillaParser is a Java Html parser based on mozilla’s html parser. it acts as a bridge from java classes to Mozilla’s classes and outputs a java Document object from a raw ( and dirty) HTML input.
• HTML Parser is a Java library used to parse HTML in either a linear or nested fashion. Primarily used for transformation or extraction, it features filters, visitors, custom tags and easy to use JavaBeans.
• NekoHTML is a simple HTML scanner and tag balancer that enables application programmers to parse HTML documents and access the information using standard XML interfaces.
• JTidy is a Java port of HTML Tidy, a HTML syntax checker and pretty printer. Like its non-Java cousin, JTidy can be used as a tool for cleaning up malformed and faulty HTML. In addition, JTidy provides a DOM interface to the document that is being processed, which effectively makes you able to use JTidy as a DOM parser for real-world HTML.
• HtmlCleaner is open-source HTML parser written in Java. From a (dirty) HTML document, HtmlCleaner reorders individual elements and produces well-formed XML. The process is like the creation Document Object Model (DOM) in browsers. But you can provide custom tag and rule set for tag filtering and balancing.
• TagSoup is a SAX-compliant parser written in Java able to parse wild or nasty HTML as found on the web. A C++ port of the library is also avalaible.

The last 3 items are more cleaning tools intended to output well-formed balanced HTML code. Anyway, as JTidy have to build a DOM object of the HTML, you may use this to elegantly access data from your raw/dirty html source.
So for the time being, I would probably focus on Jericho HTML Parser and HTML Parser who seem the best candidates for the job. Moreover, they offer documentation and samples to get quickly started. I you already tried any of these, I would be happy to hear your recommandations.

On the other hand, if you need a generic parser, I would recommend JavaCC or JParsec. And if you are not satisfied yet, you may still look on java-source.net directory.

Now, let’s also talk a bit about Groovy, I only find 2 basic articles about parsing HTML in a Groovy way:

• http://groovy.codehaus.org/Testing+Web+Applications
• http://blog.foosion.org/2008/06/09/parse-html-the-groovy-way/

… which seems quite limited, so I should probably import some of this POJA (Plain Old Java API).  Or maybe you can point me other references?

• sh404SEF: sh404SEF rewrites Joomla URL to a more user-friendly format (SEF URL or URL rewriting). It also manages title and meta tags (with or without .htaccess file). It also claims to be fully compatible with Joomfish, OpenSEF, VirtueMart, Community Builder, Docman, … and all that for small DB query time overhead! From a personal perspective, I had trouble to make it running with Virtuemart and JoomFish at the same time (but there were some leads in the forums to get it working). Check the author’s website.
• Artio JoomSEF: This component will rewrite your Joomla! URLs not to be only Search Engine Friendly, but also to be human easily understandable. It provides support for multilingual sites using JoomFish. You can also fully customize your meta tags or your 404 page. It supports the same 3rd party plugins like Joomfish, VirtueMart, Community Builder, Docman and more… Check the author’s website.
• SEF Advance: SEF Advance is a Joomla SEF component which extends its built-in Search Engine Friendly URL functionality and raises it on a higher level. But be aware that this one costs about 40€. Check the author’s website.

For Joomla 1.0 users (if there are still any alive), you can have a look to the Open-SEF extension.

You may find other useful resources in the Joomla SEF extensions. As always, I’m looking towards your comments.

• JoomSEO: JoomSEO is a Search Engine Optimizing(SEO) bot making Joomla content more Search Engine Friendly(SEF). JoomSEO dynamically creates meta keywords, meta descriptions, changes the title on the fly (Adds the sitename for Joomla 1.5), adds heading tags to content titles and more.   This seems to me a pretty good start point. Check the author’s website.
  
• Joomla 1.5 SEO Patch: This component can improve the SEO settings of Joomla 1.5 with this free SEO patch for Joomla 1.5. The patch allows you to gain fully control over the html head section, the META and title settings in the http header settings in Joomla 1.5 generated pages. Check the author’s website.
• Joomap: Joomap is a sitemap component for Joomla that shows the normal Menu Structure, Content Categories, Sections and Virtuemart Categories in a hierarchical tree or a Google Sitemaps list. It can support all your content along with localization. Check the author’s website.
• Xmap: Xmap is based on the Joomap component. It allows you to create a map of your site using the structure of your menus. Check the author’s website.

I still have to look deeper the difference between Joomap and Xmap, but I think I’ll definitely try the JoomSEO plugin.
If you want other kind of extensions related to SEO/SEF/Sitemaps, have a look to Joomla Extensions for SEO & Metada and Joomla Extensions for Sitemap.
You can also find some tutorials and informations about SEO with Joomla on http://www.joomlaseo.net.

If you have other resources or extensions, feel free to share

• Dependency Walker: http://www.dependencywalker.com/
• Process Explorer: http://www.microsoft.com/technet/sysinternals/utilities/processexplorer.mspx
• Reflector .Net: http://www.aisto.com/roeder/dotnet/
• RemoteSoft .Net explorer: http://www.remotesoft.com/dotexplorer/
• DILE (dotnet IL Editor): http://sourceforge.net/projects/dile
• Dis# : http://www.netdecompiler.com/
• Xenocode Fox http://www.xenocode.com/Products/Fox/
• Spices .Net: http://www.9rays.net/Products/Spices.Net/
• Remote Soft : http://www.remotesoft.com