Mon Dec 28, 2015 4:25 am
The wiki article on Detour Prevention Mechanisms
uses the word "Prevention" in the title and in four additional places. This choice of word has led, I believe, to widespread misunderstanding.
Waze's detour detection technology is, I'm told, not designed or intended to prevent detours
. It is meant only to penalize
detours, which may or may not result in prevention, depending on circumstances.
The word prevention
in the title and elsewhere in the article leads the naive reader to believe that Waze is designed to disallow detours at all times; that a detour that does occur reflects Waze failure. This is not accurate!
I'd like to recommend that we change the word "prevention" to "penalty", rewording around the change where appropriate. The word penalty
will much more correctly reflect the underlying technology.
The acronym BDP can continue to be used without modification, thanks to the fact that penalty
both start with the letter p
Mon Dec 28, 2015 5:23 am
This is a bit getting pedantic. All of Waze's penalties are designed to prevent unwanted behavior UNDER NORMAL CIRCUMSTANCES.
Mon Dec 28, 2015 5:34 am
I'm no fan of pedantry. But there is confusion on this topic. For example I have been told sternly that I should not say prevention "failed" in circumstances where prevention did not occur merely because the penalty was overcome.
If confusion persists under the current choice of words, then it seems worthwhile to attempt improvements. I don't think that's pedantry.
Sent from my iPhone using Tapatalk
Mon Dec 28, 2015 3:34 pm
The other quite confusing issue at this point is that the actual mechanics of whatever you want to call this are set forth here: https://wiki.waze.com/wiki/Detour_Preve ... anisms/USA
However, we have not yet been able to reach agreement on how to present this information on the global page.
Mon Dec 28, 2015 4:22 pm
This concept is already mentioned on the page with an entire dedicated paragraph.
It is important to understand that Waze uses a penalty-based routing system. It computes the total penalties for a route and selects the one with the least cumulative penalty. Therefore it is possible for big detour routes to occur when the primary route is extremely slow.
Exactly how slow = extremely slow, is proprietary information the devs do not wish to publish, but the concept remains in heavy traffic it is possible to overcome the penalty, and the prevention behaviour.
Mon Dec 28, 2015 11:06 pm
The penalty concept is indeed available from the existing documentation. However, I believe beginning and advancing editors would find the penalty concept easier to absorb if it did not seem to be contradicted by the word prevention.
As I've mentioned, I was somewhat reprimanded recently. I used the word "failure" to refer to an event in which detour prevention did not prevent a detour. Of course, I understand the point. If other factors overcome the penalty, that is not a failure.
My points are only these: (1) despite the existing documentation, confusion persists on this topic; and (2) this confusion could be reduced, especially among beginning and advancing editors, if we stopped calling this a prevention technology, since it is considered to be working perfectly even when it doesn't prevent anything.
Fri Jan 01, 2016 5:32 am
You should not have been reprimanded, at most gently corrected. A BDP failure would occur where there was circumstances should have led to the penalty disfavoring the route.
Personally I don't care if it is called a failure or anything else. The only reason to possibly care is that you want to avoid confusion between a BDP bug (the failure I described in the previous paragraph), and a BDP that has been overcome (i.e., BDP applied, but penalized route was still the best route after penalization).
Finally, to me, prevention and penalty are not contradiction. The penalty is intended to prevent these detours where they don't provide a much faster route. Nowhere does it say it ABSOLUTELY prevents this type of route. And that's more pedantics, I guess. I don't see the current language as problematic. I'm OK with changing the language, but not with changing the nomenclature, since that would be disruptive. If you can propose an article change that leaves the nomenclature intact, that's fine. The problem is, inherently, you're asking for a nomenclature change.
Mon Jan 04, 2016 3:52 pm
I really do believe that the difference between prevention
falls well beyond a pedantically split hair, but I accept that others might not.
At the end of the day, it doesn't matter much what we editors who already have some grasp of the internals think. The wiki is intended primarily to support new and advancing editors. Would they be more likely to understand the concept if we called it Big Detour Penalization
rather than Big Detour Prevention
? I suspect so, but that's only a guess.
I'd like to invite new and advancing editors to comment on which of those two conveys the concept of a penalty with intent to discourage, but not completely disallow, detours
.P.S. I'm aware the current nomenclature has become habit and will be hard to change. But it definitely won't change if the topic is never raised, so I wanted to raise it and go from there.[EDIT: clarified "end of the day" sentence and fixed error in this edit.]
Last edited by DwarfLord
on Tue Jan 05, 2016 5:20 pm, edited 2 times in total.
Mon Jan 04, 2016 5:26 pm
I'm in favor of penalty as that is in line with how things work behind the scenes and get editors using the right terms and thinking about WHY waze generated that route.
Sent from my iPhone using Tapatalk
Mon Jan 11, 2016 2:13 pm
I'll say what I said in another thread – it's called detour prevention, not detour prohibition. Like "crime prevention," it doesn't always work.
"Detour Prevention" is in the context of "Detour Prevention Mechanism" which is a mechanism designed to prevent detours—not to stop them from ever happening, but to do so within reason.
Powered by phpBB © phpBB Group.
phpBB Mobile / SEO by Artodia.