[OSM-talk] OpenStreetMap Future Look

[Frederik Ramm]

Jeff,

On 08.01.2013 19:22, Jeff Meyer wrote:
>     what kind of data would you want to see?
>
>
> Data supported by numbers, external studies, some employment of the
> scientific method that include evaluation of alternatives or the absence
> of what has been done, rather than long speculations in email.
>
> Data kind of like what you ask for here:

But your request for supporting my post through data was *before* I 
posted the message from Pawel about "without any research into this..."?

> If this is true, why are most of your reactions to suggestions
> explanations of why those suggestions are bad ideas?

I think that many suggestions would receive a much warmer welcome if 
they were worded more like a call to action and less as a complaint.

Compare:

"Hey folks, I've been thinking it would be great if there was a way to 
announce my mapping event to all mappers in the vicinity. I've looked at 
the rails port code and I have some ideas how to do it, but before I 
implement something I'd like to know if any work has been done on that 
already? Also I think I might some help in setting up my rails instance, 
is there a howto somewhere or would someone be willing to help me 
through IRC or so?"

(To which I'd probably reply, great idea, I'd like to see that, but make 
sure you implement this via some sort of opt-in mechanism because we 
have had complaints about such invitations in the past.) [*]

with:

"Hey folks, this OSM web site is a real problem with its 90s design and 
lack of social features. This is going to fall over soon if you don't 
act. I mean, even the simplest things aren't possible - I tried to 
invite mappers in the vicinity to my mapping event and it cost me three 
hours. I think this is a real problem, and it also demonstrates the 
general lack of ambition in OSM. I think you should collect more 
donations and hire developers because otherwise these problems are never 
going to get fixed."

(To which I'd probably reply that I don't think the lack of this 
particular feature is such a big deal and that all the problem talk 
helps little.)

These are hypothetical examples and nobody has said either of these 
things, but they are both essentially about the lack of a certain 
feature, and they will get mixed receptions.

But there's another thing. I've been on these lists for a very long time 
and have discussed many aspects of OSM(F) with many people. Most ideas 
brought up here are not new; most of these discussions I have been 
through lots of times. This means that I often know, or at least have a 
hunch, where a certain idea might have a weakness. Read again the 
hypothetical reply marked [*] above - it is intended as a completely 
positve response but even that contains a pointer to a problem 
("opt-in") which might give someone the impression that I was rejecting 
or criticising the idea even though I was just relaying results of 
earlier discussions or experience we had.

I have also sometimes reacted negatively to priority setting. If you say 
"it would be nice to have A" then I might say "sure, sounds good"; if 
you then continue "... therefore I want OSMF to make a plan how to 
achieve A and acquire the necessary funding" then I might say "uh, wait 
a minute, I think that the limited manpower that OSMF has at its 
disposal should perhaps not be commited to A so lightly, at least not 
without thinking if B or C would achieve more for the same effort".

This might come accross as "Fred says A is a bad idea" but in fact it 
isn't - it's just "A might not be as important as you think".

> Why not just say, "Hey, good idea. Go for it. Here's a link to the
> typical process for getting new features added"?

I'll try to do more of that.

> That said, your first reaction to the suggestion of adding routing to
> the home page is negative. Then, later, you describe one routing effort
> you've been working on as good - and - it sounds like someone's already
> made a decision to add it to OSM.

Let's not say "decision"; once you are in it for a while, you have a 
feeling for what is likely to happen. Many things in OSM are not 
decided, they are just discussed for a long time and a decision then 
evolves.

If you want to read up on past discussions regarding the "routing" 
issue, this might be a good starting point:

http://lists.openstreetmap.org/pipermail/strategic/2011-January/000215.html

http://lists.openstreetmap.org/pipermail/strategic/2011-March/000278.html

On 11th March 2011, the Stratgic working group resolved to recommend 
adding routing to the OSM site, albeit with the main vision of using as 
a debugging tool (see 
http://www.osmfoundation.org/wiki/Working_Group_Minutes/SWG_2011-03-11) 
and without recommending a specific routing engine.

On 13th April, the OSMF board discussed the issue and minuted: "SWG 
Routing policy accepted. Emilie to advise TWG and return with budget 
requirement". On 11th June 2011 the board minutes contain the sentence 
"Development is progressing in the community. At the moment there is not 
a routing engine suitable for the OSM website."

So the basic decision to have routing is there and has been there for 
almost two years; it is likely that the shortcomings of various routing 
engines (gosmore - flexible but slow; osrm - fast but inflexbile) will 
be addressed by offering a mix. I had initially provided a couple of 
extensions to the rails port that added a routing tab and a matching 
controller to pass through requests to interchangeable backends, but I 
believe Kai (apmon) as probably re-written that several times by now and 
his currrent work is here: 
https://github.com/apmon/openstreetmap-website/tree/routing2 and the 
branch can be tested here: http://routing.apis.dev.openstreetmap.org/

>     by the time we'll run it on osm.org <http://osm.org/> it will be
>     relatively mature.
>
> Criticizing an idea without revealing that you're involved with a
> similar project. What's up with that?

Huh? I hope you'll want to withdraw that accusation after you have read 
up on things. I've tried to give you a run-down of three years of 
routing discussion so that you have a background on which to judge 
issues and people but I can't do that all the time, for every topic. I'm 
sure you know that and that's why you are writing up a lot of stuff on 
the Wiki, which is good; maybe well-maintained Wiki documentation will 
help to avoid such misunderstandings in the future.

Any of the stuff I explained above can be found out from existing 
archives and documentation, it is not *just* in my head and there's 
nothing I can "reveal" that is not already out there in the open - but 
of course having been there at the time does sometimes make it easier to 
find stuff.

> Has a decision been made that that *is* the routing engine that will be
> added to OSM? If so, great. I look forward to it. Has that been
> publicized in the community?

As far as I am aware there has been no formal decision about which 
routing engine(s) to add (and which of those to operate ourselves). But 
unless something new miraculously descends from the clouds, I think that 
the gosmore+osrm combo is most likely.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"