You can thank your alpha-testers (and you can thank the USPTO)

The USPTO has, over the decades, had a bit of a spotty success rate at designing e-commerce systems.  This blog article highlights two reasons to allow yourself some guarded optimism in this area — the fact that USPTO is doing real alpha-testing of its Patentcenter system, and that a small but very active core of USPTO customers are carrying out very aggressive alpha-testing of that system.

First a reminder of what “alpha testing” and “beta testing” mean.  These terms can mean different things to different software developers.  

In many programming environments, “alpha testing” means “simulated use by the developers themselves” and means “not yet being used for production by  external users”.  In such environments, “beta testing” means “for the first time actual external users are using the system, maybe for test transactions and when we get very brave, maybe for revenue and production transactions.”

In such programming environments, “beta testing” often also means something like “the design of the underlying databases and general code flow is very nearly set in stone and what remains is bug fixes in single lines of code” while “alpha testing” means something like “the system design is still in somewhat of a preliminary stage and almost anything could change at this point”.

In such programming environments, the assumption is that the people doing the alpha testing — who are simply some of the programmers and developers themselves — are fairly confident that they know enough about what it means to be a real user that they feel they can lock in the design of the underlying databases and the general code flow before ever letting a real external user touch the system.  In other words, the first time that a real revenue customer touches the system is in beta-test.

You can read about this distinction between alpha testing and beta testing in this Wikipedia article.

My firm serves for example as a beta-tester of VOIP telephone protocols with the telephone service provider VOIP.MS as I have blogged here.

In the context of USPTO systems, the plain reality is that if USPTO were to postpone the involvement of real users until the beta-test phase, it is a recipe for failure.  The problem is that there is no way that USPTO’s own developers and programmers can actually do a good job of testing a system the same way that real USPTO customers would do.  As a striking example of this, consider the ePAVE e-filing system which USPTO tried to roll out some twenty years ago.  USPTO did not meaningfully involve revenue customers until it was too late in the design process.  Too many bad design decisions had been made by that point.  ePAVE never managed to get even a 5% penetration rate among those filing US patent applications.  USPTO eventually scrapped the entire ePAVE system, migrating to the EFS-Web system.  

One of the main points of this blog article is to recognize, with approval, USPTO’s decision to do meaningful alpha testing of its Patentcenter system.  For more than a year now, USPTO has been doing real alpha testing of this system.  Patentcenter is a system which is intended eventually to replace both Private PAIR and EFS-Web.  It will be a top to bottom replacement of both systems.  Many back-end systems, which USPTO customers normally do not see in any direct way, will also be replaced or substantially upgraded as part of this migration.  

And what I would like to illuminate brightly is that USPTO is doing the right thing with Patentcenter.  USPTO is doing real alpha testing, and importantly, is doing it with real external customers.  The point here is that USPTO, to its credit, is self-aware enough to be able to admit that its own developers and programmers are not in a position to test a system like this the same way that real revenue customers would test it.  So for more than a year now, USPTO has been running an alpha test of Patentcenter with real users who are filing real patent applications.

By now, some thousands of real patent applications have been filed in Patentcenter by dozens of real external customers.  Responses to Office Actions have been filed.  Issue Fees have been paid.  US patents have been granted on applications where every step of the process took place through Patentcenter.

During this period, the alpha testers have done their best to explore Patentcenter, to find and report problems and defects, and to carry out the work of prosecuting patent applications.  During this period, the alpha testers have seen instances where USPTO people have actually corrected problems or redesigned portions of systems to overcome design mistakes.  By this I mean that the alpha testing has actually served its purpose.

At some point the USPTO will feel confident enough in the design of Patentcenter to lock down the system design more or less completely.  At that point the only types of changes that USPTO would be able to make are “bug fixes”, for example changes of a single line of code to correct a misspelled word or something.  That is when, traditionally, the terminology of “beta testing” kicks in.

Anyway I invite members of the patent community to recognize USPTO’s doing the right thing by involving actual revenue customers in its alpha testing of Patentcenter.

And I invite members of the patent community to recognize the contributions and efforts, for more than a year now, of the alpha testers.  Our firm is one of the alpha testers of Patentcenter.  I know who many of the alpha testers are, but I do not feel at liberty to name them.  Maybe at some point there will be an opportunity to give some recognition by name to these long-suffering alpha testers of the Patentcenter system.

Are you one of the alpha testers of Patentcenter?  If you are comfortable letting people know this, please post a comment below.