Monday, April 19, 2010

Software genies

Note, the title doesn't say 'genius' as some might expect. There is a lot of hubris that has been generated by systems weenies, what with the exhibited prowess, the accumulated monies, and the wizardry aura.

Guess what, folks? Throughout our times, humans have been foil for peddlers and malfeasance. Nothing wrong with that, capitalism comes out of that realm. Too, our materialistic, and scientific, progress has caused us to ignore the fact that there is a whole lot that we don't control.

For the most part, the systems, including control, people have won the day. Certain operating systems just crash with no warning (and on a massive scale). So we grumble, but to whom can there be a complaint? Too, the web user, many times, runs into some interminable process (hey, peoples, the basic notion here is undecidability). In this case, just redoing the action can overcome the problem.

There are many, many examples. And, OS, and system, crashes have been the direct cause of billions of losses. How much is being bled, malfeasance'd, from our economy with web-assisted means?

Anyone take the blame here? With these types of losses?

---

Well, in some cases, such as medical computing, we've seen cases where culpability was established and enforced. Too, in flight systems, there is a lot of thought and effort put into safe computing. And, finally, there is going to be an effort at beefing up cyber security.

Ever think the WWW as safe? Of course, that question is directed in the context of mature reason and actual experience. The actions of a whole bunch of unawares seem to indicate that the real problems are not understood. Perhaps, training ought to be better (computing, like other important things, ought to be safe).

Now, did you ever think that computer-related failures would show up in your car? Again, 'computer-related' here subsumes a whole lot, including the systems and the mathematics. Did you think that hubris would allow the spawning of potential 'genies' on the populace on such a large scale (thank goodness, for the older technology)?

But, it has happened and will continue.

---

It's interesting that Consumer Report did some testing before the fact. One thing to be concerned about is the completeness of those things tested before a car is released to the public. Of course, 'completeness' is used advisedly (think Godel), as we cannot attain such, to the nth degree that the political types would like to have us believe.

No, there will be tradeoffs which will require us to be smart. And, the tradeoff analysis ought to be done in a visible arena.

So, we'll see failures. Is the forum open? No, and, then they 'hack' a solution and cast it out across the waters! I guess we need to make it more apparent that changes, themselves release new 'genies' in a manner that requires extreme care (like making software an industry problem, for example).

---

One thing to point out is that hubris leads many to think that some entity that is a collection of systems and sensors is equivalent to a real being. Part of the hubris rests on advances that show that control, and convergence, are partly algorithmic. Yet, that success can hide just how close we come to divergence.

The recent test results point to an area that we need to rethink. And, from this blogger's view, it'll end up requiring what might be specified as 'truth' engineering (of course, there may be a better moniker to use as a handle for the subject).

Remarks:

01/23/2015 -- Software? Well, we are talking more than apps (latest craze). We are dealing with fundamental questions which, then, gives rise to normative issues in mathematics (and, by extension, to the computational).

03/09/2013 -- Zeno can be put forth today in other than a facetious sense.

03/23/2012 -- Renewal of the idea (and related energies) via Cooper and CiE.

11/21/2010 -- Three years ago, it was said: Computational foci raise miraculous need. Still applies.

10/07/2010 -- Several principles need to be explored, such as the ergodic one.

09/28/2010 -- It nice to see the IEEE weigh in. Notice: sensors galore, driver in the loop, ...

05/10/2010 -- Out of control, essentially. Commentary, and cartoon, via USA Today is right on.

Modified: 01/23/2015

Wednesday, April 7, 2010

Who knows?

Earlier, there was a question about 'who is to know?' in the financial context. Economics is not called 'dismal' for no reason.

Then, we observed the frenzy created by the media about some car problems. Essentially, there are many assumptions about quality that go unaddressed until after the fact of accidents, it seems. Some of this may be due to a slackening of effort, since business has been driven by costs for some time now (that is, the CEO as king as opposed to making the consumer the king, okay?).

But, some problems arise from the difficulties of the cyber-physical. That is, this topic is now being addressed by the NSF, however there is much more to it than the current views cover.

It was good to see in the case of the car problems that experts who are outside the discipline, namely NASA and NAS, are being brought in as these issues go fairly deep. That action raises the level of talent being devoted to the problem and may become more necessary than not as products become more complicated.

By the way, one set of experts addresses myths vs facts. This is an interesting read as it shows that the manufacturer has to consider failure modes in order to instruct the driver what to do when an incident occurs that needs attention. Actually, problem avoidance requires identifying these modes of possible failure, too, so as to lower their possibility during design or to anticipate corrective actions, as we see here.

This, then, raises, again, the question of who represents the consumer in this case? Who would have thought that the merely rolling (as we all do when at a stop light on a hill while we move our foot from the brake to the gas pedal) would raise the likelihood of such a dire consequence?

Remarks:

02/26/2011 -- Another go.

02/08/2011 -- There was a report today concerning a study on the SUA problem that has been going on quietly. More news will be coming later when the report is technically analyzed.

08/13/2010 -- Recent news that it's user error.

04/19/2010 -- Genies, no not genius, indeed!

Modified: 02/26/2011