Web Forefront

I've seen a few stories in recent months all having to do with scaling, there's of course the Google I/O conference that covered what can be considered 'proprietary'-'non-available' technology, the Twitter scaling nightmare -- which is backed by Ruby/Rails BTW -- and yet another on how Facebook is buying thousands of servers to support its traffic -- which is backed by PHP of all things -- its no doubt an interesting topic, but what are the real technical options out-there for scaling ? This entry will cover a few case scenarios I've worked on related to scaling, elaborating on the software that makes it all happen.

Reaching a consensus is a difficult task in any endeavor, just look at technology standards were numerous parties are pulling in different directions in order to get better leverage for their 'standard' based products. One such process going on right now is the one being held by the OpenAjax alliance, which will hopefully have a say on future browser standards, but what is probably more important about this process is its openness, which sheds a real light on what the software community has to say about the techniques browsers should be supporting.

With more and more software deployed as services over the web, the infrastructure used to support it is becoming increasingly important, as a consequence data center technology is also gaining its fair share of attention. Google recently held an event showcasing some of the technology it uses in its data centers, from custom made circuit boards, custom made server racks to its own custom file system, its an impressive look at the backbone of what is probably the most scalable 'web application' on the Net.

Advertising is now part of the web's DNA, the scientific and collaborative origins of the web might still be there, but lets face it, once business sense got involved it was only a matter of time for advertising to come to the forefront. This entry will cover a few practices I've seen that revolve around the money-pit that is advertising, content, domains and registrars, seen from someone who both buys advertising on the web and tries to make revenue from advertising, if you do any of these things on the web, read on for some dirty little secrets that are not so obvious and are poised to break anyone legitimately involved in this same loop.

OSGi is slowly but surely moving into Java server-side territory and a few other SOA product lines in the industry. Among the most noted, you will find a few of the earliest Java Application Servers now being tagged 'OSGi compliant' or 'OSGi based', in addition, popular Java frameworks like Spring have also blossomed OSGi integration sub-projects. But what are the actual implications behind OSGi ? And more importantly how can you work with it ? This entry covers an article I recently wrote on the subject which includes a hands-on OSGi example, as well as the potential implications OSGi will have on the overall Java and SOA ecosystem.

In exploring Python as yet another alternative to software development, here are two links to articles I recently wrote on the subject, one is related to a Python web framework named Turbo Gears using Oracle as the backing database and the other to an interoperability mini-article exploring Python calls inside Java programs using Jython.

The Web mail market is pretty much like any other market -- two or three very well known and dominant players, with smaller one's poaching from the big one's -- the big one's in this case being Microsoft's Hotmail/MSN/Live service, Yahoo's mail service and Google's Gmail service. Well, there is something seriously wrong when you can't get through to users on one of these mail services -- email being the collaborative and communication medium it is -- so if you in some way operate your own domain and need to send legitimate email to Hotmail/MSN/Live users, read on, as it will bring to your attention what may be happening to your messages and probably save you a few hours or days of frustration.

Throughout my career I have seen two lines of thought for evaluating IT professionals, there is the 'look for talent' mantra which I would classify as the generalist search, and there is the 'sink or swim' attitude which I would classify as looking for the specialist. I've been on both sides of the stick, so it came to my surprise that there is in fact empirical research outside of computer science that establishes a link between happiness/grumpiness and generalists/specialists.