Blog/November 2013/Nov 8th

Nov 8th – no new devar torah this week, but here’s one from a few years back.

Hi, again!

We were all very busy this week. Sarah’s been hard at work at school, getting home late and exhausted. Esther and I worked on the business and did lots of stuff.

I had an MRI of my knee done very early Monday morning (1:30 in the morning), at the extreme other end of Jerusalem from us. So I went there, spent some quality-time with the late-night denizens of Jerusalem (yes, there are some), and made it back home around 03:00. Took an hour to get to sleep, then woke up as usual around five in the morning. That day at work I was pretty useless (although I did manage to get stuff done).

Then on Tuesday evening I had an appointment with a neurologist, to try and figure out whether or not there was a more serious reason for my occasional loss of peripheral vision than “visual migraines”. I was actually supposed to make that appointment three years ago when I first had the sudden vision loss, but since it went away I forgot about it. I’ve had these “episodes” several times a year, and this last time the ophthalmologist reiterated I should get a neurological exam to rule out “more serious” problems. So I went, and he said that the most likely scenario is indeed just “visual migraines”, but he was sending me for a CT scan to rule out anything more problematic. So this coming week I get scanned on the other end of my body from this week’s scan.

Then on Wednesday morning I went back to the orthopedist for a review of the MRI results. He told me that I do indeed have a “torn meniscus” in my knee, and that it won’t get better on its own. He suggested either surgery, where they remove the torn up meniscus, or doing nothing at all but “don’t overload the knee”. He actually recommended not doing the surgery at this point because “you function well” — which is a matter of opinion, but I can walk most of the time without too much pain. His suggestion was that I cut down on the walking and find other, less “loading”, exercises to perform (like swimming or stationary bicycling). Sigh.

Then on Thursday, Esther and I went to Tel Aviv to meet with our business consultant, who had put together most of a revenue forecast for us. He made suggestions for ways we might increase the revenue (a good thing) and reduce costs (also a good thing), so we have some things to consider. In the meantime, I have to revisit my original Gantt charts for my projects’ development, and update them so I can give the consultant more accurate numbers from which to make forecasts. What it really comes down to, is that getting my 8th development environment fully functional is becoming very important. And I did spend time on 8th, most of it writing documentation to help me keep in mind just exactly what I want to accomplish and to make it easier to make final documentation later. As part of this effort, we got a “Mac mini” for development work, as well as an “iPad mini” for testing. Now Esther’s trying to figure out how to use the iPad.

As I’ve mentioned before, Sarah’s being employed by us as a “QA” person. Soon that will be a more significant rôle, and we’ll need more hours than she can give us. We’re OK for now, but you might wonder what’s so important about QA? I mean, it’s the designers and developers who get the kudos, not the poor souls who check that the products work. How important is good QA after all?

Software is complex and the bigger the project, the more unforeseen interactions there may be. Here are a few examples of truly major problems which better QA (and better software-engineering practices) could have caught:

• The portal for Obamacare: healthcare.gov. There are so many problems with it that is unlikely to ever work correctly, but it is apparent that no effective QA was done at all. Hint: when you open a site which you expect millions of people to access, you really do need to check it under simulated loads.
• The Therac-25 medical radiation-therapy device. It was supposed to give bursts of X-rays to kill cancerous lesions. But a software glitch caused it to sometimes give deadly doses of radiation. In this particular case, the software was not designed cleanly enough to be tested well, and people actually died because of it.
• The Ariane 5 flight 501, which lasted less than a minute once defective software caused it to veer off its flight path. Because pre-flight tests had not been performed properly, the problems were not discovered.
• The Mars Climate Orbiter, which failed to properly enter Mars orbit because the different teams used different units of measurement (metric vs Imperial) and failed to coordinate.

Those are just a few of many examples of software failures which cost many millions of dollars. Now you know why software QA is important…

Well, Dinah did not join us last week, but B”H she will this week! So she and we will enjoy:
tomato soup, homemade ḥalla, ḥamud (meat version and veggie version), rice, cauliflower couscous, green beans, radish salad, roasted eggplant salad, vegetable chili and fruit crumble.

Until next week,
shabbat shalom!

---

Top: Blog

Prev: Nov 1st

Next: Nov 15th

---