jump to navigation

Nick Finishes The Job June 16, 2011

Posted by ismywebsite in general.
trackback

Ahh what a lovely thing to hear:

Deltik has moved onto a shared webhost and a VPS, provided by Almighty Servers.
Our previous webhost, IsMyWebsite, has pretty much committed suicide. We’re outta there.

So a suicide eh? That would imply I intentionally did something to damage/kill the service. Ahh I remember now what I did. You’re right. Unfortunately, if I didn’t go along with it I was considered unreasonable and inflexible.

Since version 5 is ‘almost done’ as it was for the last 5 weeks before I finally pulled the plug on it, why don’t you just launch it then? I’ll let you use one of our domains (say ismywebsite.info) and just see how it goes. Since I’m sure a launch was imminent.

You can expect Deltik to be much more stable and have less downtime. (No guarantees; it’s possible that I can accidentally botch up something because I’m not used to the new provider..)
Also, Deltik should load faster now. The new hosting provider outperforms our old one.

Well Mr. Liu, I wish you luck with the reliability. I’ve set up monitoring on it and only time will tell. As for loading faster, it might be worth your while to do a side by side comparison. For the record though, it still appears you’re comparing a VPS to a free hosting plan. Not only that, but for someone so opposed to the invite system, due to it restricting our services from being usable by anyone, it sounds like you have an exclusive deal with the host.

As for “it’s possible that I can accidentally botch up something because I’m not used to the new provider”, it’s not only possible but highly probable if this is your first time with a VPS especially.

Oh and get this – I heard about this from our (apparently unreliable and slow) provider WebLyte. Why? Well because of a script driving up the CPU load on your account nonetheless:

They had a script driving up the server load consistently for about 5 minutes to 2.00+

So if there’s a death here, it’s certainly not suicide.

However, there’s still a heartbeat. There are still people who believe in this site and this service, and that’s the only reason it isn’t dead. I may not see that path out of where we are, but I can tell you without a doubt that a small group of committed people can do anything. In fact, that’s about how anything gets started or happens in this world.

Nick is a very smart capable person, who did care a great deal. Unfortunately, he chose at one point to view me as an adversary, someone to be undermined and given false progress reports, and later to be opposed in a strong fashion. Since a strongly divided team is not really one at all, at that point, I had but two choices – give Nick and all the other developers the hosting entirely (which Nick didn’t even want) or take back control and do what I could to salvage things. I didn’t see any decent working relationship between me and Nick at that point. Thus, there was no other option.

But most likely Nick’s criticism will be of my blog posts here, reflecting my concerns, fears, and worries about what could happen, which not surprisingly came true in the end. You can argue if through posting, I somehow affected his team’s morale and took their courage and made them take longer to finish things. You could argue that my attempts to discuss with them what needed to be done, and try to even determine what was going on, did in fact slow them down. The fact still stands, aside from changing the password once out of frustration, they had full access the entire time and any questions about my code I answered in full.

In fact, however, if I know anything about Nick and the type of person he is, he loves a challenge, and the more people who tell him it can’t be done, the harder he works at it. It’s truly unfortunate he believed that 8 months of coding could be done in 1 month, even by a team as skilled as his, and under his devoted leadership. There was a point I realized the only thing holding him to keep working was this burning desire to prove me wrong. It wasn’t until I finally told him that I really appreciated the hard work and I wanted to support that version 5 and potentially use it, that he appeared suddenly to become very busy with school and homework and all these other commitments started getting mentioned. I kind of predicted that would happen as well. That’s when I pulled the plug.

And so Nick, the previous project I mentioned in my last blog post – everyone thinks I’m crazy to even attempt such a thing. People say it’s impossible. Nobody’s ever done it. Why – because it’s just too hard. Sessions and cookies are too complicated to figure out, let alone build a working virtual browser. Maybe you could do it – nah you’re too busy. It’s too great a challenge. There’s no possibility of success. You’ll fail. You’re not good enough. It’s probably impossible. It wont work. How would you even do it anyways? Yeah, like I said, pretty much impossible. There’s no way you could do it. You shouldn’t even try.

Good luck.

Comments»

1. Deltik - June 16, 2011

Hey, I’m featured in this blog post! 😀

2. Hostify Networks - June 16, 2011

“…the previous project I mentioned in my last blog post – everyone thinks I’m crazy to even attempt such a thing.”
No…just very misguided. It can be done, but for the amount of time you’d have to spend on it, it doesn’t even begin to make sense.

Why is the idea of making your own domain management API so appealing? Will it work that much better than what other engineers have spent months developing? Is it really worth the time required to develop? How do you plan to maintain it if you have no development team? What other parts of the site are being neglected while you’re working on this? Is this really a good use of your time? See my point?

3. ismywebsite - June 17, 2011

>> Why is the idea of making your own domain management API so appealing?

Firstly, it’s not just a domain management API. But assuming it was, this will grant complete freedom in selecting suppliers. As a result, we don’t have to pick from just Company X, Y, or Z who have an API. We could have several companies and shop the bargains.

Obviously, in a highly competitive market like domain names, many companies bargain price themselves below the actual domain registration fees. They do this at a loss, as part of a marketing ploy, hoping to regain the remainder through long term business or complimentary purchases. However, few or no bulk API domain provider would do something like this. Instead, we’d have to pay significantly more, and as a result can register significantly less domains. Since domains are important for the development of websites, and a highly desired item among webmasters.

Even factoring in thousands of domains, it’s still more cost effective to shop around and take advantage of these deals. Obviously, the equation changes as the length of the registration increases, however we currently have a very short registration length on average (hardly over a year).

>> Will it work that much better than what other engineers have spent months developing?

Probably not. However it will have a function which they cannot provide – supplier flexibility.

>> Is it really worth the time required to develop?

I believe so. Even if only for managing domains and no other purpose, it would still basically cut our costs in half. (Actually, prevent them from doubling, while allowing clients to control their domains.)

Aside from this, imagine if AdSense earnings were credited to accounts in real time – rather than once every time I have time to log into a website, download a CSV, open it in a text editor, copy it into a script, and run the script. Now imagine I could save hours in assigning channels, and make better use of the ones we have (which are limited). Even if this was only a 10% increase, that’s enough to host a heck of a lot more sites.

Next, think of the advertising networks without APIs. In fact, very few do have APIs. Imagine the frustration and headache at trying to track the earnings on all those separate services. How often would it be possible to log into 21 different sites, download information in 21 different styles, and paste it into 21 different scripts? Even if I was efficient and only took 3 minutes per service, that’s still over an hour lost. And I believe that the earnings should update in real time, once a day.

And that’s just the start. This is a framework which could be used for a lot of projects.

>> How do you plan to maintain it if you have no development team?

Well if a reasonable set of unit tests can be created, then they can be run each day and any changes to the various sites, or surprises would be forwarded to an administrator. Rather than having to manually perform all those actions, we’d just have to fix the directions from time to time. Of course, the basic underlying code would be the exact same, and only the layout/navigation path of sites would ever be expected to change. So once you have it working it’s fairly simple to keep it working.

>> What other parts of the site are being neglected while you’re working on this?

Always something is neglected. I can’t change that. And I’m not working on it at the moment. I’d like it if someone could help get it built.

4. J,R,D, Ltd - September 1, 2011

Any new developments since this?

Deltik - September 1, 2011

For IsMyWebsite, obviously not.

Please contact me for detailed information: deltik@gmx.com


Leave a comment