Forbes has done a spring review of the best of the web for 2005 and Firefox came up as their preferred web browser. Unsurprisingly, the other category that interested me was search engines and Google won that one. The list of companies, groups and organisations backing Firefox is growing at an astonishing rate. If you haven’t tried it, perhaps you should find out what they are all going on about.

Sometimes a new feature comes along that leaves you unsure if it’s a good or bad development. This is one such development. A company by the name of Remote Approach has developed a system whereby PDF files can be tagged with the addition of some code so that it reports home every time someone opens the file and reports the IP address and other details, including any unique identifiers the makers choose to add, back to the author. They are also apparently working on a method of denying access to the PDF if the reader is not online at the time.

My concern is that this could become yet another tool for tracking users habits, and also that companies will start using the facility to require users be online to read ebooks, so that they can track piracy. Since ebook readers and other such tools are unlikely to be online most of the time, this could create serious usability issues. My laptop has been very handy for reading long PDF’s while sitting on a deck chair out in the back and out of range of the Wireless network. I’d hate to lose that ability due to a restrictive new tool. I’m also not convinced that such a tool should allow collection of IP address’s and other such explicit information about users as it increases the likely hood that in future such a tool might be used by unscrupulous types. This particular tool is subscription based and as such will be under the control of Remote Approach, but there is a good likelihood that the technology can be co-opted by people of less moral fibre and that worries me somewhat. Trends seem to strongly indicate that the days of the anonymous Internet are drawing to a close. As John Bielby of Remote Approach points out, such information gathering takes place already with Web server logs, but what they don’t mention is that web users can use an anonymizer service to hide their details from web servers if they chose to do so. No such facility is currently available for the new PDF system.

We have just been offered one suggested solution (along with a $30 donation for mentioning it through December 13, 2007) to the offline use dilemma is to use a PDF to HTML Converter. This would allow you to instead use the document in straight html that would be available offline. That suggests an intriguing solution. For a long time, Google has converted pdf documents to html, and I often use that method when searching to get a sense for what is in the document because of the relatively lighter download that I have to take. Unfortunately I have not been able to test able2extract because the donation didn’t come with a copy of the software. It suggests that unlike Google, it will convert images as well as text. Google mainly converts the text portion in my experience.

I’ll admit, I have yet to actually encounter one of these files that isn’t available offline, so perhaps we are thus far tilting windmills? If it becomes common place, then I’ll definately try something like able2extract.

Third world countries have many problems to overcome. Some can be overcome by sending aid money to those countries, setting up wells, hospitals, schools and food programs, but not all of the problems can be solved in this manner. As the saying goes: “Give a person a fish and feed them for a day, teach them to fish and they will feed themselves indefinitely.” Getting these developing countries into the modern world and teaching them how to survive and indeed flourish in such an environment is the key to their prosperity. To do that you must logically begin with the school age children as they are the ones that will be in the position to take the countries forward. To that end, the founder of the MIT media lab, together with AMD, Google and News corp are putting together plans to supply a 100 dollar laptop running the Linux Open Source Operating System, to be purchased by governments and supplied to school children in the developing countries. Building a complete laptop for only 100 dollars is a serious challenge as often single parts are worth most of that amount, (like the screen for example). The likely specs for such a laptop would be somewhat lower then existing mainstream laptops but not so much that they can’t be used for most of the same things most of us use our computers for, (not counting gaming of course).

By using Open Source software like GNU/Linux, the makers avoid the heavy cost of software licenses, and as an added bonus, they avoid the vast majority of spyware and virues problems that plagues the Windows world. From an Open Source perspective, the good news is that whole new generations will learn and use Free (as in freedom) Open Source software and will become valuable additions to the OSS world once they are educated.

The units are expected to be ready for distribution by late 2006 or early 2007. Read the MIT media lab site on the project for more specific detail, or see this nice Wired article on the subject.

With the speed at which RSS feeds are growing in popularity, two things are becoming necessary. The first is ways to make money from RSS feeds, and the other is to set your server up so it can handle the ever increasing load that RSS will put on your system. This article is concerned only with the second item.

Part of the problem is that most RSS feeders seem to all fetch their feeds every hour on the hour, meaning you get hit by everybody at the same time, the other part of the problem is that many readers don’t check to see if the feed has been updated since the last download and instead just download the whole thing again. To get the most from your RSS, there are two things you should do. The first is to make sure your server uses Gzip compression like mod_gzip or mod_deflate. This will compress the XML to a fraction of its former size which will drastically reduce bandwidth and increase download speed for your users. Those Apache modules will not just improve your RSS feeds, they will improve all text based transfers (like HTML, Javascript, CSS etc).

The second thing is to ensure your feeds support “conditional GET” which basically ensures that a feed isn’t downloaded again if it hasn’t changed since the last download. Many blogs now support conditional GET, so you should check to make sure yours does also. A good choice of blog that supports both Gzip and conditional GET is WordPress, but it certainly isn’t the only one. By following the above steps, you can significantly reduce your own bandwidth bills, and also give your users a faster more pleasant experience. To check if your feed or site is using Gzip compression, you can run it though this handy tester. To help lower your peak bandwidth, you should also encourage your users to set their readers to update feeds at random times but set intervals so that they spread the load over the full hour rather then all fetching the feeds on the hour.

Apparently Windows N is not very popular with OEM providers with Dell, HP and others saying there was no incentive to sell them. The problem is two fold, firstly they cost the same as the normal Windows with Media player. The EC should have forced Microsoft to drop the price by $5 dollars for Windows N, to give OEM’s an incentive to sell them. Secondly the EC should have made arrangements for Windows N to be offered with alternative media in an easy fashion. If Realplayer (for example) could be slipstreamed onto an OEM’s install disk, then in conjunction with a small price cut there would be real incentive for OEM’s to actually consider offering it on their systems.

Put another way, say for example you were offered two cars, identical in every way, except one has power windows and the other doesn’t. If the cars are the same price, which one would you pick? To take that a step further, if the car without power windows had after market power windows that worked much the same as the normal ones, would anyone care which one they got?

Web designers these days should at a minimum look at their pages in Firefox, Internet Explorer (painful though that may be), and perhaps Opera and Safari. Often, we tend to overlook the growing number of text only browsers that happen by our sites. Tonight I happened on a great tool to help you see what your page would look like in Lynx, a graphic free browser, should you choose not to load a copy yourself. This tool allows you to see your page like it would appear to a Lynx browser.

That site offers a number of other great tools as well, including:
– a header viewer — see what your server is putting out
– a http request scan — see what you are giving servers such as your browser string, referrer, etc.
– search engine simulator — think like a search engine when looking at your pages
– backwards compatibility check — see what the page looks like with some tag types removed that weren’t supported

While none of these things are “rocket science” in scope, it is still a handly little group of tools for the kit.

According to Netcraft’s latest report, the Open Source Apache web server software has continued to push down Microsoft’s web server market share. Apache now has a tiny bit less then 70% of the worlds web server market stitched up, compared to 20.45% and dropping for Microsoft’s web server offerings. Apache gained market share slightly more then what Microsoft has lost, indicating that much of Microsoft’s lost marketshare is probably going straight to Apache. The percentages don’t sound like much until you note that the survey is based on the responses of 62,286,451 sites. Also of note is that there has been an increase of 1.7 million hostnames in the last month. Microsoft doesn’t have a monopoly everywhere it seems.