Software. Efficiency. Scalability.

Lodsys story is generating a lot of buzz on the net.

People argue that the whole software patent system is screwed up. Personally, I tend to agree that copyright is enough in software as, generally speaking, the implementation of the idea is much more complex than the idea. Think app store — it’s one thing to come up with an idea of it and quite another thing to create a viable platform, create libraries, attract developers, go through the approval process, etc.

The other part of the problem is that software patents are supposed to protect a small inventor against bigger players that can copy his or her work. Well, that doesn’t work too well because enforcing a perfectly valid patent against a larger corporation can cost millions of dollars in litigation costs. It’s quite cost prohibitive and there are just not so many cases of this. Also, most of the big players on the market have huge patent portfolios and can always counter-sue either increasing the cost of litigation or making it a wash. There are just not so many of the successful examples of where a small guy got protected by the patent system.

Having said that, I don’t believe for a second the government will have the audacity to abolish the software patent system. Even suspension of granting new software patents while the system is kept intact would be too big of a pull to swallow. However, there are certainly ways the government could deal with the issue.

A big part of the problem is granting crazy patents such as MacroSolve’s one. USPTO is swarmed with patent application and has a multi year backlog. It doesn’t have enough of the qualified personal to evaluate all the patents properly and compare them against the prior art. In this case, the solution could be crowd-sourcing. There are a lot of interested parties out there that would be happy to help USPTO with the research and try to invalidate rogue applications before they become full fledged patents.

There are plenty of concerned organizations and people and this process can be designed to make a financial sense for large companies and concerned groups to voluntarily participate in the process and provide USPTO with their research on prior art and validity of patents.

Another thing the government could do is to prohibit several areas of software from being patentable. User interfaces, file formats, protocols, things like that. Explicitly prohibit patenting different data types: if you communicate a piece of data over the network and store it on the server it doesn’t matter if it’s a geo-location or a credit card charge. Has anyone send information from a phone to the server and stored it there? Then it’s a prior art.

Overall, an open process with some adjustments to what’s patentable and what’s not could result in much better patent system. I have a feeling that guys like Lodsys will eventually generate enough will in the community to break the status quo and push for a reform.

I decided to give Google Apps a try for my domain, opletayev.com. I wanted to have a central place for my family, mostly for the calendar purposes to map all kind of dates that I always forget. My cousin’s children birthday, for instance. If my cousin reads this: I do remember their names!

So, long story short, I went and signed up for Google Apps. I got my email, calendar, and docs setup. I was a very happy panda, for a short amount of time. But I am also a nerd, so I wanted my calendar to be accessible at calendar.opletayev.com. I mean, I have my own domain so it makes sense, right?

I went to godaddy.com and realized that I have my DNS mapped to WordPress name servers, which is nice and dandy- it dynamically maps your domain name to your blog, it has Google MX integration, but it doesn’t allow you to add CNAME records. Which is, of course, what I needed to make my freshly baked Google Apps work the way I wanted.

I changed my DNS back to godaddy.com name servers. I created MX and CNAME records for google apps and I mapped @ to the IP address of my blog. Currently, it seems that my blog is up but all custom CNAME records point to the blog, not to the Google Apps site, which is puzzling.

I guess I’ll let it sit out for a while and grab something to eat while trying to figure out how to make it work.

I really like Google Buzz. I’ve been a user of Google “Leaky” Talk, for years now and I have most of my friends and co-workers on it, which makes my following/followers list instantly populated with people I communicate the most.

No message length limitation is always good. 140 characters limit on Twitter always felt artificial. The ability to attach photos and links is something I’ve been looking forward to as well. Overall, it seems like a good quality product that will gain some momentum eventually. Twitter’s Fail Whale is pretty annoying, especially when something big is going on, like the Super Bowl.

http://google.com/profiles/opletayev#buzz

Why does an instant messenger application need 8.5 times more memory than Visual Studio 2008 with 5 projects open?

I understand that Google can hide behind the “beta” tag as long as they want, but this memory leak has been around for at least 3 years already. Is it really that hard to fix it?

I guess I have to take the remarks I made in the last post back. EtherPad.com is opened to the public once again, and they are planning to open source their AppJet engine:

We have begun planning how to open source the code to EtherPad and the underlying AppJet Web Framework. We will continue maintaining new pad creation from the EtherPad home page at least until we have open sourced the code, and work hard to make sure there will be no or minimal service disruption in the future.
(Source: http://etherpad.com/ep/blog/posts/etherpad-back-online-until-open-sourced)

I guess Google realized the whole “no more users” thing didn’t look so good for their reputation, so they went back and fixed it. On the other hand, I still think they just wanted the team anyway, so why not open source the AppJet engine? It may not fit the Google software stack very well, or Google may just have a superior platform altogether.

Congratulations to AppJet and etherpad.com team!

EtherPad was a great product and it is nice to see the team being rewarded for their hard work. I wish they published more information on their AppJet engine, but it doesn’t seem that it will ever happen.

What is interesting is that right after they announced the acquisition they immediately stopped accepting new customers and creating new free pads. It basically means that Google is not particularly interested in their customers or their user base. All they wanted was the team.

Or, maybe, they wanted to get a competitor product out of the market to clear up the way for Google Wave. But, nah, I mean, Google is not Evil and wouldn’t throw money to shut down the competition, am I right? :)