Hari's Corner

Any member of any online technical community will tell you that the one question that gets asked on a fairly regular basis is the question: help me choose a web host or where can I find cheap/free web hosting?

Both these questions come up fairly often and the answer often is a brusque "google it!" Well, out of personal experience I can assure you that finding a good, reliable shared web hosting service on the web can be a fairly nerve-wracking experience and finding one that suits your pocket as well as your requirements can take quite a while, especially when you see so many sites that offer "cheap" or "absolutely FREE" web hosting.

This is where it gets really challenging to sort out the chaff from the wheat. Inevitably most of these web hosting providers do not display a full listing of features available on their hosting package and they allow the customer to "assume" quite a bit which might turn out to be false in the end. Consider the fact that one web hosting service that I found on the web offered a whopping 1000 MB of space for as low as $10 per month, but the moment you click on their feature listing, it strikes you that all is not as sweet as it seems. In a quiet corner, you find the words:

MySQL databases: 1 MySQL database storage: 15 MB

In essence, what this means is that if your website is a dynamically driven database size (which typically require a lot more space than normal, static HTML driven sites) you get only 15 MB of space to work with. Too bad about the remaining space: you can upload all the pictures and photos you want, but none of that space is "smart" space: space that counts towards making your site what it is: a dynamic content-driven website.

Paradoxically, it is the content-driven, dynamic sites that require a hell of a lot more disk space than the static ones for fairly obvious reasons. Databases aren't static entities. You can work with them: add, remove, delete, update content and this is what this generation of websites are all about. Dynamic rather than static. The web hosting provider essentially says: "you can take all the disk space you want, provided they're not used as database storage space. Have fun!" But since you don't need all that space but you need the extra database space which comes with a more expensive package, you end up paying the full amount for a fraction of the usage. In short, "you're screwed."

And what's worse, many of these hosting providers don't provide an iota of information about their true connectivity speeds, their server capacity or their uptime to maintenance ratios. Most of the genuine information is usually buried beneath heaps of unimportant details or even simply absent. Of course you should use their "contact form" and clarify all this, but then how many of us do it? It's just too much hassle to send an e-mail and then wait for a response for a few hours or in some cases, days. Many people just go ahead.

While one might argue that new or basic customers would not require the dynamic database-driven features, it is a wrong assumption. While an inexperienced user would not know what a MySQL database is or what server-side scripting implies, he sure wants to start an online club or forum, a blog or a even large content-management system. All these applications require a combination of database/server side scripting. And all these users stand to lose more when they realize one day that their disk space has run out and what the hell! They've paid for 1000 MB! Of course, it's that 15 MB of database space that has run out but then they saw the space advertised on the front page and jumped on the deal.

That is why it pays to do research. Finding a good hosting service that provides you a honest deal takes a lot of time, research and questioning. If you cannot find these answers on the website at least take the time to ask some questions like:

1. What is the total speed of your connectivity? How many servers do you provide and how do you share it?
2. What kind of domain/subdomain limits do I have? What is the cost of registering a domain name through you? How many subdomains am I entitled to on my account?
3. What is your server uptime normally? How frequently and how long can I expect downtimes?
4. Do you provide database support? If so, how many databases and how much space of it can I use in my total hosting account space?
5. What server side scripting languages support do you provide?
6. What server do you have? Is it Apache/Linux/Unix or is it a Windows server? What kind of backups are taken (if any) and how frequently?
7. What kind of front-end you provide for me to access my hosting account? (most professional hosting packages should come with cPanel at least and full FTP support).
8. Please provide me with a full list of your terms and conditions of use.

It's not an exhaustive list, but I think these should do to begin with. Of course it's our duty to do our research before paying the money. Since there are a dazzling array of choices on the table it is wise to consider all the choices carefully before making a final decision.

Remember that it pays to take your time to find out. Otherwise you may find some nasty hidden thorns along the way! Happy hosting!

I must admit that I was nervously awaiting the result of the software patents voting in the EU parliament and I heave a huge sigh of relief and satisfaction at this result.

I just read the story here.

It is with great relief that I note that for now this new law has been rejected. That vested interests have lost and common wisdom has prevailed. For software patents would have signalled the beginning of the end for FOSS and innovation in the software line. It was well known that the large corporations supported this bill because they had the most to gain from it while small companies, individuals and independent programmers would have suffered badly as a result of restrictions on software ideas. The very concept of patenting software is restrictive and such an implementation would have effectively put an end to the smaller players in the software industry. It is good to know that people at the EU had enough common sense to put an end to it.

But all is not over. I still believe that danger still hangs over our head and it requires constant vigilance on the part of every concerned citizen. Let us wait and watch developments in this issue in the coming months. But for now, all I can say is: Phew!!

Another edition of the LiteraryForums.org newsletter! And this comes at a very auspicious time indeed for we've seen some good activity in the month of June. It has been a good period for LiteraryForums and I'm glad to be back after a period of illness. We've had quite a few new members join up as well and this is quite an encouraging sign.

Changes? The most obvious change is the change in colours and design. To an extent, this was prompted by the unfortunate loss of data which ruined all my good work earlier. I was truly disappointed that a lot of painstaking work was lost because I hadn't taken any backups. Now I have decided to make some changes anyway although these new changes aren't as comprehensive as the changes I had made and planned to implement earlier. However, changes were long overdue and I'm glad to have made some of them at this point. On a minor note, there has been a forum security upgrade as well.

Another aspect is the forum look and feel. It has been my desire from the very beginning to try and change as much of it as possible to give LiteraryForums.org a unique touch and in this regard I have decided that the forum images need complete replacement. Needless to say, this will take some time, but the effort will be worth it. The advantage here is that the new images will match the new theme quite well.

On a general note, I hope that this is just the beginning of all the activity! There was a perceptible increase in forum activity last month and I hope this continues. With more users submitting original content, reviews and discussions, LiteraryForums.org can only become richer. And we're getting closer to reaching 100 members! A significant milestone!

A big 'thank you' to all those members who have made this possible!

Yours Sincerely,

Hari (Forum Host)

I had earlier posted my comparison between Debian and Slackware (see this article). All I left out in the comparison was covered by another appreciative member of LinuxQuestions.org, imitheos, who wrote this very interesting comparison and filled up most of the gaps left by me. I am much obliged to this member for granting me permission to use his post in my blog. Here are the main points comparing Debian and Slackware. I have reformatted the post for readability, but essentially the material has been pasted verbatim. Again, many thanks!

1) Package Manager Slackware has pkgtools. They are some shell scripts that untar the .tgz files and also write the contents of the package in /var/log/packages. It doesn't handle dependencies.

Debian has apt-get/dpkg I have used Slackware, Mandrake, Suse, Redhat, (Free/Net/Open) BSD, AtheOS, DGUX, Solaris, CRUX, YellowDog It is imo the best package manager that exists in the unix world. Many will say that RPM is good. First RPM doesn't handle dependencies well. For the majority of People who use Redhat/Mandrake/Suse/etc RPM is good For people like me who mess with the OS to see how it works, i have broke RPM many times and even render my system unusable (glibc )

I have yet to break apt/dpkg. Whatever i do it simply works. Gentoo's Portage/Emerge is also very good and handles dependencies, but apt is 10 laps ahead in F1 terminology.

1 point to Debian

2) RC System Slackware uses BSD init scripts. That is some scripts in /etc/rc.d Debian Uses SystemV init scripts. That is the scripts are in /etc/init.d and there are symbolic links of them in /etc/rcX.d for the X runlevel.

SystemV are not difficult and with the rc-update and some RC editors, they are very easy, but I (my opinion) prefer Slackware's

1 point to Slackware (Subjective Point)

3) Completeness Debian is a complete os (especially good for a desktop system) for example. You install the "elvis,vim" packages in both Slackware,Debian. In Slackware vi symlinks points to elvis and if you want you change it to vim. Debian has the "alternatives" system. Thatis /usr/bin/vi -> /etc/alternatives/vi -> /usr/bin/elvis There is the "update-alternatives" command which changes the symlink to which one you like. Very good solution imo.

Also, the scripts of every program are very complete. For example, sendmail' script (/etc/init.d/sendmail) even checks if i have a dialup connection and when it is up it configures the sendmail.cf with my new dynamic hostname.

1 point to Debian.

4) Dependencies As i said, Debian honours dependencies while Slackware does not. This good both good and bad.

The good of dependency checking is that when you check a program in Debian all its dependencies are calculated automatically and installed too. With Slackware you must find what they are and install them (from the website or Readme or ./configure ) The bad thing is that sometimes you get to install some things that are not necessary. For example, when i tried to install mjpegtools it needed the libdv library which is not absolutely necessary for the program to work and in Slackware i didn't compile it. This is again because Debian wants to be complete so it enables all the dependencies of each program when compiling.

I guess 1 point to Debian for most people (although i prefer to have an option like Slackware gives me)

5) Point of view Debian is a complete os (have i said it again ? ). It has an awful large amount of packages (15490 according to Debian Website, but more if you count the "unofficial" packages in www.apt-get.org)

Slackware is a minimalistic os with the point of view that i give you a simple os and you install only what you want from there (NetBSD is the same)

1 point to Debian for desktop system. 1 point to Slackware for server system (Also subjective. Some may disagree)

6) Installation Difficulty and Time Both use simple ncurses interfaces with description and everything so they are very easy to install. (I don't know why many people say they are hard)

Debian Installer is more complete and it has many translations (Although many computer terms are always english it feels good to install in your native language) Debian takes more time to install because of the huge amount of packages you have to choose from (for a custom installation)

7) Packages Update Debian is updated very frequently (Testing/Unstable). Slackware was being updated very frequently too,but now due to Pat's health reasons it has fallen behind a little.

Slackware tends to be among the first distros to include something. Debian wants to be stable than to be bleeding edge, so some packages take a little longer to be included. (Also, for a package to be included it must be stable for all architectures is mentioned in the policy i think, so that takes time) for example Debian Sarge uses Xfree86 4.3.0 while Slackware uses Xorg 6.8.2

1 point to Slackware

I had originally intended the second part of this article to cover the differences between Gentoo and Debian, but recently I had posted a comparison between Slackware and Debian at LinuxQuestions.org in response to a thread and I thought I should post it here as well, since a kind member responded appreciatively to it. It's rare when you get such praise in a competitive, highly knowledgeable community like LQ, so I thought it would be a good idea to blog it. Feel free to leave your own comments and thoughts as well. I'd be very grateful if you could.

Here it is:

If you are big into the "manual configure" thing and you don't mind the down-and-dirty, often repetitive tasks of installing dependencies by hand, editing configuration files manually and so on, use Slackware.

Slackware is the "hacker" oriented approach to Linux. It's clean and is tailored to allow you manual configuration of your system. GUI tools are minimal for most configuration tasks. There are 3d party package manager tools for Slackware, but your success will be 50-50 with automated tools.

It's logical approach can be easily learnt though, though you might start wearing of installing dependencies by hand as your system builds up

With Debian, you get convenience and power (if you choose to use it). Apt package management will give you 99.99% success if you choose to use only the official repositories and are relatively conservative the chances of breaking your system are nil.

As with Slackware though, system administration can be done manually, but you do have the choice of installing *a lot* of tools from the huge repositories of Debian and this might help to administer a Debian system more easily than Slackware.

Debian == Easier to maintain, upgrade and manage. Suited for general purpose power users with a wide range of needs. If you are the kind of user who wants to constantly keep installing/uinstalling/upgrading/removing software from your system, choose Debian.

Slackware == Suited for tweaking and learning, can be more involved when you upgrade. Suited if you don't need too many applications and if you aren't going to constantly install/upgrade/delete applications. Tailored to be lightweight, configurable and yet powerful.

Both are excellent mainstream distributions each focussed in different areas and each following a different set of priorities and philosophies.

Shopping for computer hardware and peripherals can be a pretty daunting task for most computer users because of the sheer choice of products and the wide range of prices. And this process can be especially daunting for users of Linux. Let me assure you, however, that researching a product before you buy it can pay rich dividends. I can speak from personal experience.

Google can be your best friend when researching for hardware compatibility. I urge everybody who uses Linux to do their homework before purchasing a piece of hardware. And that brings me to the main point of this article: why do hardware manufacturers continue ignoring the Linux end of the market and why they continue to respond so poorly to repeated appeals from the Linux community to provide better support and 100% compatibility with their products?

And believe me, there is no reason why hardware manufacturers cannot provide Linux drivers. Their excuses are pretty lame and dismal. When drivers for Microsoft® Windows® are provided, why should Linux be given shabby, stepmotherly treatment? And yet you have those people who have the audacity to complain that Linux doesn't support their hardware! What nonsense! What utter rubbish! While the best brains of the Linux community are doing their best to get those miserable pieces of hardware to work under Linux by long hours of toil, effort and reverse-engineering, you have people screaming that Linux doesn't work! Ridiculous.

Whatever the reasons for certain hardware manufacturers to turn a blind eye to Linux (and I won't go into that debate here), it is the duty of every Linux user (from newbie to guru) to demand their rights before shelling out the cash. It can be incredibly frustrating to ask your local store dealer/supplier about the technical issues relating to compatibility with Linux (most of them will simply blink and stare at you when you mention the word "Linux"), but there is such a thing called the world wide web. Use it to your best advantage and do extensive research before buying a product. Companies that refuse to support Linux or even allow other independent developers to create drivers for their products can suffer by losing your support. Who cares whether they're the biggest company on the planet? Let self-interest rule your decisions regarding purchasing their products.

Finally, I would like to ask you to sign The Linux Printer Driver Petition targetted at printer manufacturers who continue ignoring Linux. I found the link when I was browsing LinuxQuestions.org today. It won't take a minute of your time so I urge you to read it and sign it.

Please do put your signature on that petition for whatever it's worth. At the very least it will make sure that the voice of the lone individual will not get lost in the wilderness.