Friday, September 29, 2006

Choosing a web host

Choosing a host is a bit of a nightmare. If you have a recommendation from someone you trust, then that's a good starting point. Otherwise the problems you face are an overwhelming array of choices, and a market flooded with advertising/ misinformation/ unscrupulous companies. Here are a few do's and dont's:

  1. Don't pay any attention to "top 100" type webhost listing sites. Its all advertising - companies just pay to get on these lists.
  2. Do look around for genuine comments from users on forums etc. If you find a lot of negative comments about a particular host, stay clear. A lot of positive comments is good, but be aware that some of these could be fake comments. Use judgement.
  3. Don't focus exclusively on the numbers that hosts advertise. You might be tempted to look for the highest hosting/traffic limits to cost ratio (i.e. biggest bang for your buck!). This is exactly what many hosts are counting on - they unrealistically inflate their numbers to attract customers. 5000 GB traffic for $5 a month could not actually be supported by a host.
  4. Do look for a decent server uptime (99.5% +).
  5. Do run a whois on the host. Has the company been around for long? If not, you might be increasing your risk. Many webhosts come and go in a short time - its a competitive market.
  6. Do test out a prospective host's support. Give them a call/ email and ask them something about hosting - how long did they take to respond and were they helpful?
  7. Do consider what you will really need. How much traffic do you anticipate? How much space do you need? Remember that with most hosts you can upgrade to a higher package if traffic etc exceeds your initial expectations. Also consider what server-side support you want (php, ssi, whatever...). Dont forget to check out what versions the host is running.
  8. Do some more research. Search for "choosing a web host" or similar - read some articles (e.g. how to choose a web host).

Most of all, look for a company that you think is honest and reliable. It's much more important to find a host that will respond to your support requests, accurately charge you for services, and stay in business, than a host that offers high traffic and storage limits for low cost.

I personally ended up going with hostgator. I chose them for 2 main reasons - positive reviews on forums, and the fact that they have some sensible advice about choosing a webhost here (some of which is repeated in the above). I can't say if they are the best company out there, but so far I am happy, and would recommend them. They provide good site stats, and my html and php stuff worked first time. I have a couple of issues using their web-based email but that's relatively minor.

Thursday, September 28, 2006

Blogsoldiers: A partial revision

After receiving 3 comments and 1 email today from blogsoldiers surfers (thanks guys) I have to at least partially retract my previous post. So it seems that there are soldiers out there who actually read other people's blogs, while they are building up credit for their own...

I still suspect that there are users who dont read the other blogs. One reason for this is that a small number of blogs showed up multiple times after I had surfed only 50-100 sites using blogsoldiers. These bloggers must surely either buy credits or sit there clicking and not reading other people's blogs. However, out of about 25 referrals today from blogsoldiers, I got 4 responses. That's actually not a bad conversion rate. Of course, the title of my post was deliberately provocative - I claimed that blogsoldiers didn't work knowing that I would have blogsoldiers traffic directed to me. Would I (or others) get this many (not that 4 is that many!) different responders to less provocative posts?

A more important question is probably whether it is worth the investment of time to browse other people's blogs in order to have some traffic directed at your own. Would it be more beneficial to focus on writing good quality content, or to advertise in other ways? If you already have a decent readership, then a focus on content makes sense. But what about when you are trying to build up a readership?

I dont have any definite answers here, but I can say that blogsoldiers does indeed generate traffic, with at least some of the readers being prepared to take the time to read and respond to your content. Its probably a reasonable option if you dont have many readers, or if you enjoy randomly browsing other people's blogs.

Wednesday, September 27, 2006

An aside: Does Blog Soldiers work?

In short, I would have to say no. First of all, what is Blog Soldiers? Its a way of driving "readers" to your blog. From their homepage:

"You view other member's Blogs and in return other members view your Blog. Every 2 member Blogs you visit, 1 member will view your Blog, all Free."

Sounds good huh? Although I had reservations, I decided I should check it out. So what happens is you sign up for a free account, then start viewing other member's blogs. You have to stay on their blog for a minimum of 20s before an option to browse to another blog appears. In return, you are guaranteed that for every 2 blogs you view, someone else will view your blog. Easy traffic you might say? Yes, easy traffic. But low quality traffic. The problem of course is that the people who are viewing your blog are really just doing so in order to increase traffic to their own blog. I tried to surf with an open mind, with the intention that if I visited an interesting blog I would read it, leave a comment, or maybe even bookmark it. But I found very little that was of interest to me. That's not to say all the blogs were necessarily bad (some were), but that they weren't my cup of tea. But I suspect that most blog soldier surfers dont even do this - they simply sit there clicking on the 'next' button whenever the 20s timer runs down, without even reading the blogs. All the while increasing their own credits so that they can get a bunch of people not reading their blog!

Perhaps I'm being overly pessimistic. But my own experience was that I did indeed generate hits to this very blog from Blog Soldiers, but that not a single soldier left a comment or browsed to squidsquid.com from here - indicating that the readers can't have been very engaged. Of course I might just have a genuinely uninteresting blog, but I hope instead that most of the soldiers weren't really interested from the outset.

If anyone else has had a different experience with Blog Soldiers, I'd be interested to know...

Monday, September 25, 2006

Domain name registration

Domain name registration is easy and cheap. First of course, you have to decide what domain name you want. Today this can be a problem - a great many domain names you might think of are already taken. Some of these are taken by legitimate sites, but annoyingly many sites may be instances of domain parking (or variants like cybersquatting or typosquatting). Still, you have to come up with something - I chose squidsquid because it seemed catchy (a lot of variants like giantsquid.com, megasquid.com, bluesquid.com etc are already in use).

Many hosts offer "free" registration as part of their hosting package. However, you should consider registering your domain name independantly. This is because if you let your host do it, you cannot be sure what information they will enter in the WHOIS. For all practical purposes, whoever's information is in the whois owns that domain (you can check who owns any domain by running a whois and entering checkdomain.com). If your host enters their own information they might pressure you to stay with them, or there might be complications if they go out of business. So for piece of mind, I recommend registering your domain name yourself. Then when you sign up with a host, you just need to change the DNS servers to point at your host, which is pretty easy.

There are many specialist companies for domain registration - I ended up going with namecheap. It cost only $8.88 USD for 1 year, the process was quick and easy and I have no complaints.

Sunday, September 24, 2006

Installing development tools

Hi all,

The next step I took after coming up with my initial website ideas was to install some stuff on my machine that would allow me to start building a local version of my site. Note that all the programs I have used so far are freely available, and that I'm running Windows XP. So, what am I using:

1. Programmer's notepad. I love this program. Its a free, open source text editor. It recognises tons of files types (css, html, php etc) and displays formatting appropriate to the file type. All my files are hand-coded using this program.

2. Apache web server, version 2.2.3. Apache is open source and the most widely used web server on the internet. I used this excellent guide on how to install the apache web server on windows. Followed the instructions with no installation problems. The only problem I have at present is that I can't work out how to view the locally installed manual files! I think I'm missing something obvious here - maybe someone can help. If I open one of the html files in the manual directory (C:\Program Files\Apache Software Foundation\Apache2.2\manual) the following is all that displays in my browser (both firefox and IE):

URI: caching.html.en Content-Language: en Content-type: text/html; charset=ISO-8859-1

Seems like the manual is offered in different languages but I don't know why it doesn't just show the english version for me (I assume it would try to use the browser's language setting). Well, no big deal anyway, I can always find manual information online if necessary.

3. Php 4.4.4. Again I followed the excellent tutorial on thesitewizard How to install and configure php4 to run with apache on windows. I ran into a problem regarding compatiblity of apache 2.2.3 and php 4.4.4 on windows, which was not covered by the tutorial. Found a fix after some searching (sorry I can't remember exactly where!) but it involved using some different dll's (one is php4apache2.dll I think) and a couple of small configuration changes. Also had to download the Microsoft Visual c++ redistributable package so the dll's would run. After that I browsed to http://localhost/phpinfo.php and it worked! Very pleased to have a web server with scripting language installed and running on my machine - with a minimum of fuss.

4. The other thing I needed was an FTP client - for uploading to the host when I had one. I'm using LeechFTP. It's an older FTP client and doesn't have all the new features or nice interface of some newer ones. On the other hand it's free and does the job for the relatively small amount of use I give it. The only real problem I've noticed is that if you upload a new version of an existing file, sometimes the result is that you get a strange mixture of the old and new file on the server! Anyone else come across this problem? I actually seem to remember the same thing happening with a different FTP client in the past. My workaround is to ensure that I delete the file on the server before uploading the new version. If anyone can recommend a free alternative to Leech FTP, I'd be interested to know.

That's it - a total of 4 programs installed on a windows xp laptop and I have a functioning development environment. Next entry will talk about choosing a host and domain name registration.

Friday, September 22, 2006

Initial Concept

You must have heard some success stories of people making money from websites (and blogs) through google adsense content-targetted advertising (there are alternatives to adsense, but it seems to be the current market leader so I'll be sticking to it for now). After hearing such stories myself I had to wonder if I could do it. I have a background in programming, a little website development experience and also a lot of (mostly academic) writing experience. So it seemed like something I might be able to do...

After deciding I wanted to set up a site, the next question of course is what would be the purpose and content of the website? I decided I wanted some interactive content (so I could try my hand at writing some scripts), and also that it would be a comedy/humor kind of thing. I also have an idea for a more standard services-type site, but that's on hold for the time being. After coming up with a few disparate ideas, I hit on a theme. The site is best described by visiting it, but in short it is written from the perspective of a giant squid who is communicating with humans via the internet. Ahem. Yes - you read that correctly. The humor won't be for everyone, but I'm happy with it right now. Regardless of the content - you can see that at present it is a small site with simple design, with a few server side (php) scripts.

My only real advice at this stage of the process (and all subsequent stages) is to write down every half-decent idea when you have it. It doen't matter if you don't have a way draw them all together or to use them immediately.

Next I will turn from initial concept to the first steps of implementation.

Introduction

My website has been up and running for about a week. Check it out at www.squidsquid.com. So far I've had about 100 visitors. I have quite a few plans for extensions and new content etc. The next few blog entries will cover what I did to get the site where it is now. Once I've caught up with that the blog entries will be current - covering site changes/additions, amount of traffic (and revenue - if any!), and promotion activities etc.