Tag Archives: Internet

The Network N00b

I am a network n00b.  I remember when networking on Windows (and DOS for that matter!) was fiendishly difficult and I am truly glad those days are behind us.  Although it is not really related, I was reminded of those days recently when I was trying to determine why my home internet connection would sporadically drop out.

I used these drop-outs as motivation to finally replace my ADSL modem / router with a new one.  I have wanted one for a while now, but couldn’t justify replacing a working one.  The world has finite resources after all, and we really don’t need the extra land-fill!  The drop-outs commonly took the form of firstly losing the VPN connection to work followed by extraordinarily long times to resolve web addresses.

The only way I had found to correct the problem was to power-cycle the ADSL modem.  Once I had bought and installed the new ADSL modem / router, I was horrified to discover the problem had seemingly become worse!  Now, a power-cycle was not always sufficient to recover from the problem.

Fortunately, diagnosis of the problem became much simpler with the new modem.  Once “the problem” occurred, I discovered that I could still ping the gateway machine, but I could not ping the primary or secondary DNS servers of my ISP.  The new modem has a less cryptic web interface.  This was able to tell me diagnostic information such as line attenuation and signal-to-noise ratios.  (The old one probably could do this, but I had a bad “hunt-to-peck” ratio – clicking on random links before I found the page I was looking for!)

Armed with a few statistics, I turned to the Internet for possible answers (when it was available!) It did not take long to find an answer and I am annoyed with myself for not starting the problem resolution here!  Recently, our local exchange had upgraded from ADSL to ADSL2+.  I had upgraded the firmware in my old ADSL modem/router for this change but did not upgrade my line filter.  The solution to my problem was to replace my existing splitter box and line filter with a new combined splitter/ADSL2+ filter.  Since then, things have been going swimmingly!

In the end, I probably did not need to replace the ADSL modem.  But I did want some features that my old modem did not have.  Also, the old modem had a quiet high pitched whistle which I’m glad to be rid of.  You live and learn!

The problem with the Internet (Part 2)

As I have mentioned earlier, one thing the Internet does well is allow people to collaborate.  With the now common-place Web 2.0 sites, there is no end to the social interactions available on-line. 

The problem here, is not all things work well “on-line”.  Asynchronous or not, the reliance of AJAX applications to communicate with a Web server can be a right pain.  This is probably a more pronounced issue in Australia.  We are a long way from the US and an even further distance to the UK and Europe.  There is around a 200-300 msec hit taken for data to cross the Pacific.  Unless there is some major break throughs in quantum physics, this delay is simply unavoidable.  Put simply, even things moving at the speed of light, take a notable time to traverse this distance.  Three hundred milliseconds may not sound like a lot.  Rest assured, you notice it.  Visiting US developers certainly can’t believe the slight delay that happens on every web request (presuming the web server is not located in Australia of course) and conversely most Australians abroad can’t believe just how much more responsive the Internet is in the US or the UK. 

This delay is hardly anyone’s fault, of course.  But, unless you experience this delay yourself, you are not likely to ever take it into consideration in building your application.  Here in-lies “the rub”.  Web applications are not speedy.  Oh yes, in AJAX land where sunshine and rainbows occur in green fields of joy things are much better than they once were – forms are much more “dynamic” etc, but the performance is still woeful compared to the previous generation of desktop applications.

This gives rise to the second issue.  No-one wants to develop desktop applications anymore.  The computer industry is at least as guilty as the fashion industry for being swept along with the “latest craze”. If you aren’t developing web-based applications (regardless of whether they are practical, or in any way benefit from their onlineness*) you aren’t in demand as a programmer. 

Do customers care about the technology used in the applications that they use?  Of course not – as long as the application does what they need it to do.  This lack of caring should allow the industry to stick with tried and trusted technology, rather than invent new technologies.  However, the inverse ends up being true.  That is, because the customers don’t care what technology is used, they don’t end up demanding that the industry use tools that work best for whatever the problem is at hand.   

Sometimes the Internet is the right tool for the job.  Unfortunately, the trend has been to bash it into the only tool for all software jobs.  And that leaves those that use the tools with a less optimal solution than they may have otherwise had.

* I made this term up to suit my needs… but the Internet assures me I wasn’t the first to do so!

The problem with the Internet (Part 1)

One bad segue deserves anotherWelcome to my first post on the new site. What will happen to rolypolycat? It will still exist, but become the typical “family photo album” type web-site found everywhere on the Internet under obscure URLs. You don’t have to be terribly astute to note that all the old blog postings I made on the site, I’ve transferred over to nerdrider. But (and this is a terrible segue) what if I hadn’t? What if I’d simply abandoned the old web-site to gather virtual dust? The information I’d presented there, would be locked in time. – Both web-sites are hosted on a paid server. The domain names are paid for. So if I simply didn’t renew my account with the vendor, you could imagine that eventually, the hard-drive space would be reclaimed, the DNS entries removed and a small chapter (more like a “generic sentence summarising many aspiring bloggers”) of Internet history would come to a close. But sometimes it appears to be cheaper for a vendor to buy more hard drive space, than reclaim old disk space.

I value my time very highly. I don’t object to anyone who realises its real price. Time is one of those things you can’t buy and only have a limited supply of. IT staff only have a limited time to deal with every issue that comes their way. If they haven’t had the time to set up proper practices, they won’t be using their time effectively. If a hard drive fills, then you have the two choices of “empty it” or “replace it with a bigger one”. Economically, if you spend too long working out what can be removed from a drive, it would be cheaper to replace it. In a world where the global consciousness is finally realising that resources are limited, it’s almost criminal to suggest replacing working hardware is the right choice. But, in the real world, it happens. A classic example: Eighteen months ago, I changed ISP. As is fairly standard, both old and new ISPs offered “free web-space” for a personal web-site. On my old ISP’s website, I had a PHP page with phpInfo() and another page with pictures of strobe ants (that I thought may have been fire-ants). No – I’m not the sort of nerd that studies insects, rather I was concerned that we had fire-ants but the DPI web-site didn’t have the facility to upload photos for them. So, free web-space seemed like an easy solution. Some eighteen months later, my first foray into web-sites still exists.

The computer industry travels at an amazing pace. The technology used today wasn’t here ten years ago. Maybe some research boffin was developing the ideas / software / hardware, but for the most part, it simply didn’t exist. “Best practices” come and go as standards evolve. In our society, we increasingly turn to the Internet as a source of knowledge. We expect that we may have to be wary about the truthfulness of things stated, but we don’t often wonder if its factual basis is still matching any advancements in the field we are studying. Compare this with the Internet’s predecessor AKA a library… Take a book off the shelf, and you always get a feel for how old the information is. Dog-eared pages, faded text, a battered cover, all help convey a sense of age. If you take an engineering guideline off the shelf and see it was published in 1930, you may expect that what you read will be outdated. The cues of age can be far more subtle on a web-page. Without an actual date being displayed on a page, subtelties in fashion and trend (fonts / colours / images etc) will most likely be your only guide. Match that with the pace that the computer industry moves at, and you can see a problem utilising materials found on the Internet as the latest and greatest thinking in the computing field.

I guess my points are:

  • If you are a consumer of Internet material, always look for supporting evidence to indicate what you are reading is correct and up to date. It is relatively cheap to put together a web page. And, you don’t need credentials or a reputable publisher behind you. Without prior writing experience or credentials, try getting a book published in dead-tree form!
    If you are a provider of Internet material, always date your work. It’s current now, but it doesn’t mean it will be by the time someone reads it…

Add to Technorati Favorites