JOSM rendering improvements the last days - I'm guilty!

classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|

JOSM rendering improvements the last days - I'm guilty!

Ulf Lamping
Hi!

As some people using the latest JOSM versions already noted, I'm
currently (again) trying to improve the JOSM rendering output to be
even more useful to our valuable mappers ;-)


So if you update JOSM to the latest nightly build, you'll get:

- more tags now shows an icon: amenity, shop, sport, ...
- more tags are dashed now: tracktype, aerialway, boundary, ...
- ... and lot's of other changes!

Most of these changes are based on the tagwatch output - BTW: thanks
for this great tool from my side!


But the most remarkable change I've done is to add some rules for
stuff that *shouldn't* be used IMHO.

Most notably, tags intended for ways not to be used on nodes:

- lot's of highways (e.g. highway=motorway doesn't make a lot of sense
on a node)
- lot's of waterways
- lot's of railways
- oneway (makes *no* sense on a node)

There are other stuff like oneway=no that also doesn't make a lot of
sense as this is already the default. Also some stuff that was used
times ago but there are better ways to tag them now (e.g.
highway=viaduct vs. bridge=viaduct) are marked.

Think of this as a very basic "data consistency check"!



All in all: If you find a "no parking sign on a bright yellow ground"
or a bright red dashed line you probably may rethink your tag -
especially if you're a newbie ;-)

BTW: As this is a new way to display those tags, they will show up
lot's of times on already well tagged areas (one mapper in cyprus
already noticed this).




Although I've tested the changes in several different areas, I'm not
saying that any of the above is the final truth, so think of it as a
"public beta test".
If any of the above changes turns out to be really counter-productive
(like the already reverted rendering of junction=roundabout), we have
to think about better ways.



Hopefully this makes mapping with JOSM even easier as before and in
the long term improve lot's of strange database entries ...

Regards, ULFL


P.S: Please *don't* send any mails like "I want to have rendered this
and that" or "it would be better to ...". I already know a lot of
stuff that still could be improved with JOSM rendering myself, thank
you. Be nice and you'll probably get some more improvements ;-)

_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/josm-dev
Reply | Threaded
Open this post in threaded view
|

Re: JOSM rendering improvements the last days - I'm guilty!

Bodo Meissner
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ulf Lamping wrote, on 15.07.2008 01:00:

| But the most remarkable change I've done is to add some rules for
| stuff that *shouldn't* be used IMHO.
[...]

| All in all: If you find a "no parking sign on a bright yellow ground"
| or a bright red dashed line you probably may rethink your tag -

It is a good idea to do these checks. It helped me to find some nodes
with highway tags that probably have been added by accidentally adding
the node to a set of selected way segments.

But IMHO it would be better to add these checks to the validator plugin,
not to the rendering mechanism.
Reasons:
It is better to get a list of elements with errors/warnings than looking
around for these signs. In areas with many objects I have to zoom in to
see the signs, so I have to scroll around to find all these errors.
Instead of showind an icon only, the validator plugin could display a
message telling what exactly is wrong with the tags.
I can disable some of the checks in the validator plugin if I don't like
the warnings.


Bodo
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkh8WhYACgkQnMz9fgzDSqfzmwCfYpuU/pxzjsDtqciA4N1gLZ/o
D60AnRbii7vCN+ztAFqevqnwaKP7tioz
=z71e
-----END PGP SIGNATURE-----

_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/josm-dev
Reply | Threaded
Open this post in threaded view
|

Re: JOSM rendering improvements the last days - I'm guilty!

Petr Nejedly
Bodo Meissner napsal(a):

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Ulf Lamping wrote, on 15.07.2008 01:00:
>
> | But the most remarkable change I've done is to add some rules for
> | stuff that *shouldn't* be used IMHO.
> [...]
>
> | All in all: If you find a "no parking sign on a bright yellow ground"
> | or a bright red dashed line you probably may rethink your tag -
>
> It is a good idea to do these checks. It helped me to find some nodes
> with highway tags that probably have been added by accidentally adding
> the node to a set of selected way segments.
>
> But IMHO it would be better to add these checks to the validator plugin,
> not to the rendering mechanism.
Disagreed (to some degree). While the checks should certainly be in the
validator, seeing them offhand without running a tool, especially as soon
as you make a mistake is very valuable too.
(On the other hand, having validator plugin immediatelly check just
created/modified entities would solve most of this).

--
Petr "Nenik" Nejedly, NetBeans/Sun Microsystems, http://www.netbeans.org
355/113 -- Not the famous irrational number PI, but an incredible simulation!

_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/josm-dev
Reply | Threaded
Open this post in threaded view
|

Re: JOSM rendering improvements the last days - I'm guilty!

Dirk Stöcker
In reply to this post by Ulf Lamping
Hello,

> Most of these changes are based on the tagwatch output - BTW: thanks
> for this great tool from my side!

You know http://tagwatch.stoecker.eu/ ? I was very hard-working to set
that up.

> But the most remarkable change I've done is to add some rules for
> stuff that *shouldn't* be used IMHO.
>
> Most notably, tags intended for ways not to be used on nodes:
>
> - lot's of highways (e.g. highway=motorway doesn't make a lot of sense
> on a node)
> - lot's of waterways
> - lot's of railways
> - oneway (makes *no* sense on a node)
>
> There are other stuff like oneway=no that also doesn't make a lot of
> sense as this is already the default. Also some stuff that was used
> times ago but there are better ways to tag them now (e.g.
> highway=viaduct vs. bridge=viaduct) are marked.

Some notes:

a) Would it be useful to add a new key to condition?

You now do 3 times (yes, true, 1)
        <condition k="oneway" v="yes"/>

I would suggest a
        <condition k="oneway" b="yes"/>
where b (boolean) catches all boolean values.

Helpful? If yes, I would implement it.

b) For the illegal tags you check: Could you make a list? I want to set up
the tag checker validator with a pretty easy list like this:

node: highway  = *
*:    bridge   = no
way:  railway  = level_crossing

General idea: "type:tag=value" of wrong elements

(Better suggestions welcome :-)

Maybe also using regular expressions like
node: /.*way/ = *        /* (sorry have no better idea :-) */

Ciao
--
http://www.dstoecker.eu/ (PGP key available)

_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/josm-dev
Reply | Threaded
Open this post in threaded view
|

Re: JOSM rendering improvements the last days - I'm guilty!

Maarten Deen
> But the most remarkable change I've done is to add some rules for
> stuff that *shouldn't* be used IMHO.
>
> Most notably, tags intended for ways not to be used on nodes:
>
> - lot's of highways (e.g. highway=motorway doesn't make a lot of sense
> on a node)
> - lot's of waterways
> - lot's of railways

You do know that there are waterways and railways tags that are intended
for nodes?
E.g. railway=station, waterway=lock_gate

--
God's Final Message to His Creation:
We Apologise for the Inconvenience
(Douglas Adams)

_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/josm-dev
Reply | Threaded
Open this post in threaded view
|

Re: JOSM rendering improvements the last days - I'm guilty!

Dirk Stöcker
In reply to this post by Dirk Stöcker
On Wed, 23 Jul 2008, Ulf Lamping wrote:

>>  You know http://tagwatch.stoecker.eu/ ? I was very hard-working to set
>>  that up.
>>
> Yes, I think it's really good to have a daily "overview".
>
> But: If I open http://tagwatch.stoecker.eu/Europe/En/keystats_shop.html, I
> don't get a list of shops, but only a *, which is not very helpful! Is that
> intentional or did someone changed the Wiki?

I do not use the wiki for ignores, but an automatic detection (with 200
entries max ATM). The "shop=" exceeds these 200.

See http://tagwatch.stoecker.eu/Europe/En/tags.html

There is a link to have a look at the first 200 found entries (not the 200
most encountered!):

http://tagwatch.stoecker.eu/Europe/En/ignored_shop.html

If you fix some of the wrong texts, maybe the limit no longer catches. Or
I need to increase the limit.

> Well, first of all: Don't call them illegal "in public" - I'm already
> wondering that you didn't get a friendly reminder from Frederik that there's
> no such thing as an illegal tag ;-)

:-)

> P.S: We already have maplint and the JOSM Validator plugin. Is there a need
> for a third way or would this only add to confusion?

Oh, probably that is unclear. This will be part of the JOSM validator.

Ciao
--
http://www.dstoecker.eu/ (PGP key available)

_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/josm-dev
Reply | Threaded
Open this post in threaded view
|

Re: JOSM rendering improvements the last days - I'm guilty!

Dirk Stöcker
In reply to this post by Dirk Stöcker
On Wed, 23 Jul 2008, Ulf Lamping wrote:

>>  a) Would it be useful to add a new key to condition?
>>
>>  You now do 3 times (yes, true, 1)
>>          <condition k="oneway" v="yes"/>
>>
>>  I would suggest a
>>          <condition k="oneway" b="yes"/>
>>  where b (boolean) catches all boolean values.
>>
>>  Helpful? If yes, I would implement it.
>>
> Sounds reasonable.

Ok. Now two new options exist:
a) replace k= by b= to make a boolean test (like also done e.g. in preset
tester and other places).
b) remove k= and b= totally to test against existence of the key.

I changed the first part of the styles/standard/elemstyles.xml to
introduce this new features, but left the remaining parts unmodified (I
don't want to break anything :-)

Ciao
--
http://www.dstoecker.eu/ (PGP key available)

_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/josm-dev
Reply | Threaded
Open this post in threaded view
|

Re: JOSM rendering improvements the last days - I'm guilty!

Ulf Lamping
Dirk Stöcker schrieb:

> On Wed, 23 Jul 2008, Ulf Lamping wrote:
>
>  
>>>  a) Would it be useful to add a new key to condition?
>>>
>>>  You now do 3 times (yes, true, 1)
>>>          <condition k="oneway" v="yes"/>
>>>
>>>  I would suggest a
>>>          <condition k="oneway" b="yes"/>
>>>  where b (boolean) catches all boolean values.
>>>
>>>  Helpful? If yes, I would implement it.
>>>
>>>      
>> Sounds reasonable.
>>    
>
> Ok. Now two new options exist:
> a) replace k= by b= to make a boolean test (like also done e.g. in preset
> tester and other places).
> b) remove k= and b= totally to test against existence of the key.
>
> I changed the first part of the styles/standard/elemstyles.xml to
> introduce this new features, but left the remaining parts unmodified (I
> don't want to break anything :-)
>
>  
a) You mean replace v by b, right? That would be the way to go IMHO.

b) You mean remove k and v totally, right? I don't like that thing, as
the existance of the key and if it's yes, true, 1, ... is really a
different thing.

Regards, ULFL


_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/josm-dev
Reply | Threaded
Open this post in threaded view
|

Re: JOSM rendering improvements the last days - I'm guilty!

Dirk Stöcker
On Mon, 4 Aug 2008, Ulf Lamping wrote:

>> Ok. Now two new options exist:
>> a) replace k= by b= to make a boolean test (like also done e.g. in preset
>> tester and other places).
>> b) remove k= and b= totally to test against existence of the key.
>>
>> I changed the first part of the styles/standard/elemstyles.xml to
>> introduce this new features, but left the remaining parts unmodified (I
>> don't want to break anything :-)
>>
>>
> a) You mean replace v by b, right? That would be the way to go IMHO.

Yes.

> b) You mean remove k and v totally, right? I don't like that thing, as
> the existance of the key and if it's yes, true, 1, ... is really a
> different thing.

Yes. It is. That's why I implemented both :)

When you replace v by b, you do a boolean test (means essentially all
boolean type values are handled as "yes" or "no". Non-boolean values are
handled as they are -- thus ... k="highway" b="secondary" ... would work
also (as secondary is not boolean), but this is only a side effect and
should not be used.

When you remove v or b totally, you test for existence of a key. This will
be useful in seldom cases (I found one -- oneway attribute for nodes :-)

Ciao
--
http://www.dstoecker.eu/ (PGP key available)

_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/josm-dev
Reply | Threaded
Open this post in threaded view
|

Re: JOSM rendering improvements the last days - I'm guilty!

Ulf Lamping
Dirk Stöcker schrieb:

> On Mon, 4 Aug 2008, Ulf Lamping wrote:
>
>>> Ok. Now two new options exist:
>>> a) replace k= by b= to make a boolean test (like also done e.g. in preset
>>> tester and other places).
>>> b) remove k= and b= totally to test against existence of the key.
>>>
>>> I changed the first part of the styles/standard/elemstyles.xml to
>>> introduce this new features, but left the remaining parts unmodified (I
>>> don't want to break anything :-)
>>>
>>>
>> a) You mean replace v by b, right? That would be the way to go IMHO.
>
> Yes.
>
>> b) You mean remove k and v totally, right? I don't like that thing, as
>> the existance of the key and if it's yes, true, 1, ... is really a
>> different thing.
>
> Yes. It is. That's why I implemented both :)
>
> When you replace v by b, you do a boolean test (means essentially all
> boolean type values are handled as "yes" or "no". Non-boolean values are
> handled as they are -- thus ... k="highway" b="secondary" ... would work
> also (as secondary is not boolean), but this is only a side effect and
> should not be used.
>
> When you remove v or b totally, you test for existence of a key. This will
> be useful in seldom cases (I found one -- oneway attribute for nodes :-)
>
Ah, missunderstanding on my side. I thought you meant to *replace* the
former way, and you thought about *adding* it - which makes sense at
least for oneway on nodes ;-)

However, I'm unsure if b) will be used often and this functionality
makes reading the rules at least a bit more complicated.

Anyway, if it's done already I just don't mind :-)

Regards, ULFL

_______________________________________________
josm-dev mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/josm-dev