Re: Map of Postgresql Users (OT)

I took a look at your map, this is great wonderful that you were
already thinking along these lines.

Looks like there are two issues here: (1) Getting the data points for
users/developers maps, and (2) using the most appropriate mapping
technology.

I think we can use the Zeemaps service (http://www.zeemaps.com) for
number (1) above. The way this works is that you create a map and set a
few passwords: (a) moderator, (b) member, and (c) viewer. Using the
moderator password, you can set certain map properties, e.g.,
attributes to maintain with each entry, color legends, etc. With a
member password you can modify entries and their attributes, and with a
viewer password you can just see details without the ability to modify
anything. A casual user can just see the map with the markers and their
names, without access to any details for the entries. Hence
distributing the member password to the group and having each person
add his or her details would be a good way to go. The number of points
for Postgresql users can, of course, go pretty high. So far, we have
seen reasonable performance with up to 500-600 markers. I don't know if
there is an easy way of partitioning the user base into segments that
can fit within this range. Another idea would be to just split the map
after say 500 points.

For the number (2) the choice seems to be between Google Maps vs.
Mapserver and its associated Open Source tools. Having looked at
Brent's follow up emails, it seems that Mapserver and associated
toolkits can give much more flexibility. Plus they have the advantage
that we are not limited by Google's terms, the most annoying of which
is that you cannot save, copy, or distribute the map images created.
So, it seems best to move towards the Open Source alternatives,
specially since they have better images as Brent points out. I've been
looking at some of the Mapserver toolkits, and there seems to be an
overwhelming amount of choice available for what to pick and not. So, I
don't feel qualified to pick the right technologies to use for the base
implementation. If someone can pick the mantle on that, I'd be happy to
work with that person to see how we can exchange information from the
map points stored from (1). If for the time being you feel comfortable
continuing with Google Maps, then there's multiple ways we can share
information developed by (1) through some kind of a web service API.

Claire

On Oct 25, 2005, at 2:33 PM, Robert Treat wrote:

> Actually I have already made a basic google map for the website[1]
> meant
> to replace the old developers map. Check out the blog post at
> http://people.planetpostgresql.org/xzilla/index.php?/archives/76-
> Maptastic.html
>
> I'd be interested in talking more about getting more content into the
> map and getting it in officially; One concern I had was finding a way
> to
> include information that won't bog down end users with too much data.
> (I've noticed large numbers of points tend to do that)
>
>
> Robert Treat
>
>
> On Tue, 2005-10-25 at 12:11, Claire McLister wrote:
>> Thanks.
>>
>> I looked at the Mapserver maps, and at first glance it seems Google
>> Maps API provides better map images and more interactive features
>> (zooming, panning, JS popups, satellite/map views, etc.)
>>
>> So, what would be the advantage of Mapserver apart from it being based
>> on Open Source?
>>
>> We use Postgresql server to store the data. Currently, we do not use
>> PostGIS as we do not do much geographical queries. Over time we could
>> move in that direction to offer better geographical query support.
>>
>> Posting this (or a more detailed map) on the Postgresql web site
>> sounds
>> like a great idea. We could even allow people to add markers for their
>> locations, and even enter more information about themselves to help
>> postgresql users connect with each other. One potential opportunity is
>> for users to quickly find consultants in their own area.
>>
>> How would we initiate this activity?
>>
>> Best wishes
>>
>> Claire
>>
>> On Oct 24, 2005, at 2:28 PM, Brent Wood wrote:
>>
>>> How about using Postgres/PostGIS & UMN mapserver for a fully Open
>>> Source
>>> software & Postgres based map?
>>>
>>> For a map like the one at
>>> http://www.qgis.org/index.php?option=com_wrapper&Itemid=53
>>>
>>> perhaps on the Postgres web site?
>>
>>
>> ---------------------------(end of
>> broadcast)---------------------------
>> TIP 6: explain analyze is your friend
>
> --
> Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL
>
>
> ---------------------------(end of
> broadcast)---------------------------
> TIP 9: In versions below 8.0, the planner will ignore your desire to
> choose an index scan if your joining column's datatypes do not
> match