Hacker News Reader

Comment by peey | Hacker News Reader

This is comical. It patents the entire system from HTML forms to get the input to prepare the SQL query, to server sending data as formatted HTML back to client.

This has been the bread and butter of PHP for two decades. I'm glad no one tried to enforce this to take away like 99% of the internet.

View on HN

I wouldn't hold up in any serious suit, but it could be a good patent trolling weapon. Let's hope that when IBM inevitably goes the way of the cuckoo, this does not fall in the wrong hand and/or patent reform has happened by them (fat chance imho).

It's already expired, so no danger at all.

Chances are they obtained about 500 patents of similar reach per year. This is an arms race, with companies aiming to be able to countersue any patent infringement with hundreds of patent infringements the other way.

You're completely correct. Every big software company encourages their developers to apply for patents on everything they can think of, using cash bonuses and recognition. It's absolutely an arms race, because any company that didn't do this would be standing naked if a patent troll decided to sue. It really demonstrates the ridiculousness of the patent system and how far away it is from the original intentions (even assuming those original intentions are worthwhile, which is dubious).

Patent trolls don't produce anything that can infringe on a patent, so having a patent portfolio does nothing to protect against them.

A patent portfolio helps when a competitor claims their patent is being infringed. A company with a large patent portfolio can cross-license patents and avoid a lawsuit.

If that patent was filed in 98, it expired in 2018.

I wonder if lots of bad parents have expired or are expiring soon? I remember reading the most dumb patents on slashdot in the early 2000s so maybe.

I guess we can all thank the original dot-com boom for creating a bunch of patents a little over 20 years ago! :P

Software patents have gotten a lot harder to get with a lot more stringent a set of requirements than at the start of the dot com boom, at least as of the Alice ruling in 2014. It could always be reversed, but for now, most of these really bad patents are going to die out/be invalidated and hopefully reduce the power of patent trolls

Irrelevant nitpick: I believe the saying you’re after is ”going the way of the dodo”, referring to the fact that dodo birds are extinct. Cuckoos are still extant.

Although you can also ”go cuckoo”, as in crazy.

Haha you're right, thanks! Something was rubbing me the wrong way as I wrote it but I couldn't figure it out.

Just because a patent had been granted doesn't mean you can enforce it. But you might be able to pressure companies into paying anyway because a law suite would be more expensive even if they win.

The US (and other countries) law are missing IMHO two parts:

- really serve penalties for abusing the patent systems

- serve penalties for abusing the copyright systems

This has been the bread and butter of PHP for two decades

And many other things too, but someone had to be first. I doubt it was IBM but there would have been some elapsed time between TBL inventing HTML and the first time someone dynamically generated it from an RDBMS. That’s the person who could claim prior art.

PHP didn't use a common namespace (a customizable intervening macro file used to map keywords between the form fields and DB2 SQL).

I did exactly this in 1996 when we implemented a case management system with a web-based frontend. It used a home-grown webserver written in Perl that was talking SQL to a Sybase instance.

And it was definitely not something unique. This project started because I was asking (probably on Usenet) for advice on what GUI framework we should use, and was recommended to build the application using web technologies.

The fact that IBM managed to get a patent on this is just one more of an almost infinite number of examples showing why the patent system is broken.

Perl that was talking SQL to a Sybase instance

I remember Sybperl, since everyone was using Perl for CGI in the 90’s anyway it was a no-brainer to use it for that. Sybperl was released in 1990!

Yes, and after sybperl, DBD::Sybase/DBI became all the rage in the late 90s. The IBM patent is obsceleted at this point, though I still try to see how many lines of code I need to do same thing with perl/sybperl/dbi on other platforms using other languages such scala/mysql.

Did you use an intermediate macro language and a per application macro file that mapped fields from the HTML to SQL on per request basis?

(This patent does not claim presenting SQL result in HTML -- it claims a particular way of mapping between the POST HTML and DB2 SQL using a customized macro language -- referred to as a common namespace in claim 1)

Reply on news.ycombinator.com