there is a fine line between agile and fragile
Back in April VMware hosted the 2007 Technical Solutions Exchange (TSX) for the EMEA region in Nice, France.
After a small delay, the presentations from the event has now been made available for download.
Seems like it was a good event, sadly I didn't have a chance to attend it. Somehow I did manage to get myself mentioned though, as Robert Dell’Immagine (Director of Community over at VMware) had a list of sites he wanted to mention:
Too funny.
Sadly the server that powers this site as been down since the afternoon on Friday the 26th of January until now. The reason is basically a bad raid controller that needs replacement (there will be photos later...), and I'm waiting for a new one now. How long the server manages to stay up this time I don't know, hopefully it'll manage to stay up a while.
The downtime also affects my email, so I might not be able to reply to any emails until the problems has been solved. Don't worry though, all incoming emails are cached and stored at an external site and delivered to me as soon as my server is responding again.
Most of the time self-hosting is fun, but this proves not to be one of those times.
Today I realized that a bunch of the photos in my Gallery 2 was missing. To be more precise, the full size image was missing but all the derivatives (thumbnails and resizes) were displaying just fine. Since I normally don't browse my own Gallery viewing the original full size images this problem had not caught my attention until one of my users informed me of it.
I promptly enabled debug mode, and discovered that Gallery was complaining that it could not find the original file on my filesystem. Upon further digging, I discovered that for some reason filename case had not been retained when I moved the g2data folder from my old server to the new one, converting them all to lower case. And, to make things even worse, it had not done this for all files, just some of them. Why this case-conversion happened while I was manually copying the files between the servers I have no idea. It certainly wasn't the fault of Gallery 2, it expects filenames to be consistent and not change around behind it's back.
My initial thought was to dig up HDD from my old server and re-transfer the files, but that would have taken considerable time as my g2data directory is about 15GB big. Trying to manually identify which photos were affected by this would be a horrendously tedious task too, but I needed to find a way to deal with it somehow.
Thankfully Bharat came to the rescue. After I explained the problem to him, he whipped up a 80 line PHP script that walked through my entire Gallery 2 database, comparing filenames in the database with filenames on the filesystem and identifying the ones that had a lowercase version present. Then the script would spit out a complete list of the whole mv command that had to be executed to fix the problem at hand.
The script came up with a list of 880 images that had to be renamed, which would have taken me hours trying to figure out manually. All I had to do was to copy the output of the script into a bash script and execute it, and hey presto, problem solved.
This is a perfect example that knowing developers is good for you. I urge you all to go and hug your favorite developer today. They are definately worth it.
A couple of days ago, Chris J. Davis spilled the beans about a new, and currently under heavy development, blogging package called Habari. I immediately gave it a test run, and so far it does look like it might be going somewhere, and that fast.
Interestingly enough, I read a posting on neosmart.net, called "How WordPress Spoils Developers". Interesting read, but clearly filed in the fanboy category and it's not exactly unbiased post, but it does bring up one question that I think needs an answer.
In a nutshell, the blogging market is c.l.o.s.e.d. – as in no more room, and most importantly, no more competition; because let’s face it, whatever you’ve got, it’s just never going to be good enough.
Now, is this true? Should we all just abandon everything because someone else already has figured out the best way to do something, and there is no way to top that? If that was the case, we would see no progress at all in any area. This doesn't just apply to computers and software, but just about everything. It's all been done to perfection, move on, there is nothing to see here. Do we believe so? Ok, I'm probably going a bit over the top here, but still. If we applied this kind of thinking to Web Browsers, we would never have seen Firefox and probably never seen Internet Explorer 7 either.
When I think about it, we wouldn't have seen Wordpress either. After all, when Matt forked b2/cafeklog and started on Wordpress, Moveable Type was the blogging platform everyone used and was widely regarded as a good product. The Wordpress about page explains why Wordpress has been developed, and if we applied the same kind of logic to its history it would never have seen the light of day.
Sorry, I just don't buy it. Wordpress isn't perfection, the only code I've seen the really looks like poetry is the Gallery 2 source code. Thats probably because I was never really good at analyzing poems in school, but thats an entirely different matter.
As far as my experience with the Wordpress developers/community goes, the biggest advantage it has are the non-official developers that are devoting lots of time on plugins and support. In fact, the best selling point for Wordpress is the community it offers, not the main development. Of course the main development give the opportunities that the plugin developers take advantage of, I'll give them that.
Now, back to the real reason why I'm posting this. Habari. So far Chris J. Davis and Khaled Abou Alfa is out of the closet, but given that I know who the rest of the crew is my firm belief is that Wordpress will be given a run for it's money. I'll even promise to do whatever I can to help it's development and rise to stardom.
As mentioned a few days ago, I got an offer via reviewme.com and I had some ethical concerns regarding it. Basically I decided I would give it a try, and then see how I felt when I was done.
This never happened though, as it turned out that the buyers of the review wanted me to review a physical product, not a website. This meant that the product would have to be shipped to me, from the US to Norway, which pretty much makes it impossible for me to complete the review in the stipulated 48h. I contacted reviewme about this, and got the following reply:
Our system can't support reviews of actual products -- it is designed so that bloggers can review websites, designs and content.
I did, however, get contacted directly by the people that wanted a review and subsequently made a deal with them. I said no to the payment, but agreed to review the product if I got a copy. So, at least I'm home free regarding payment for reviews and I really don't think that I will take up any offers through reviewme again.
The thought process this has stimulated made me realize that I really shouldn't get paid for reviews, but if I were to review a product I should at least get a copy of the product to play with. Reviewing products I haven't tested would make no sense after all.