Wiki Edit War on using/avoiding semicolon lists

classic Classic list List threaded Threaded
124 messages Options
1234567
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Richard Fairhurst
jgpacker wrote:
> [8 lines of text plus 240 lines of quote]

Repeated request:

Please snip the message to which you are replying to reduce it to the minimum required.

Anyone continuing to post disproportionately formatted messages like this will be removed from the list.

Thank you.

Richard
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

jgpacker
    Better try to query for "13" in ref="3;10;13;113;133" without loosing your sanity.
    Next day I will add ref="3;10;13;113;133;13E" — will you update your query?

I believe the following regular expression is enough for both examples:
    ref ~ "\b13\b"

\b means word boundary (any character that starts or ends a word, such as space, colon, semicolon, etc)

However, word boundaries can be slow and are not recommended if you need to search large areas (e.g. whole world, germany or similar).
In this case, we could use something like:

    ref ~ "(;|^)13(;|$)"

which can be read as: either semicolon or the start of the value, followed by 13, followed by either semicolon or the end of the value.
I would recommend to also allow a space before 13 (because people sometimes like to add an extra space after the semicolon), making it:

    ref ~ "(;|^| )13(;|$)"


Note: For technical reasons, if you need to use a word boundary in Overpass QL, use \\b (you need to escape the backslash character). This is not the case in Overpass XML.

PS: I don't know Perl and don't want to learn it. Regular expression is a common feature in mature programming languages.
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Marc Gemis
In reply to this post by Richard Fairhurst

Please snip the message to which you are replying to reduce it to the
minimum required.

The problem might be that some mail programs (such a gmail), just reduce the included message to a button with 3 dots. So, one might not be aware that there are hundreds of lines under it.

regards

m. 


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Ineiev
In reply to this post by Никита
On Wed, Jan 21, 2015 at 11:55:19AM +0300, Никита wrote:
> > Just because one can use a regular expression to grep out a certain
> meaning doesn't mean it's a good thing to do and will always work
> We easily revert these edits in Russia.

I'm sorry, I don't believe you are entitled to speak for the whole
Russian comunity. in fact, you have the same communication issues
with Russian mappers as with this mailing list.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Richard Mann-2
In reply to this post by Marc Gemis
Click on the dots, ctrl-a, delete. It's a lot easier than regex.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

jgpacker
I agree.
Sorry, I thought the previous messages we renecessary so the server could find out the who answered to who and to which thread this belongs.
So when you said "snip" the message, I thought of quoting the part you are answering and not of excluding previous emails.

2015-01-21 12:51 GMT-02:00 Richard Mann-2 [via GIS] <[hidden email]>:
Click on the dots, ctrl-a, delete. It's a lot easier than regex.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging



If you reply to this email, your message will be added to the discussion below:
http://gis.19327.n5.nabble.com/Wiki-Edit-War-on-using-avoiding-semicolon-lists-tp5830523p5830843.html
To unsubscribe from Wiki Edit War on using/avoiding semicolon lists, click here.
NAML

Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Никита
\b means word boundary (any character that starts or ends a word, such as
space, colon, semicolon, etc)
However, word boundaries can be slow and are not recommended if you need to
search large areas (e.g. whole world, germany or similar).
In this case, we could use something like:
    ref ~ "(;|^)13(;|$)"

Well I don't to discuss regexes actually. I learn nothing from your words. It will not help newbies at all. There lot of complexity in OSM already, starting from GPS collection to complicated tags and their poor documentation at wiki. I see no reason to complicate things with introducing any king of regex. Regular mappers get nothing from regex you wrote. Dollar means currency is USA, nothing else. "1;2;4;5;6;7;8" - just find missing "3" here, with any regex. I will not continue discussion about it with you actually.

Regexes miss set logic, regex miss default and missing values. As you mentioned they require processing power. It will eat all of your cpu. If you don't trust me just ask any DB admin is it good idea to query values in database using regexes.

I'm sorry, I don't believe you are entitled to speak for the whole
Russian comunity. in fact, you have the same communication issues
with Russian mappers as with this mailing list.
Ad hominem. So what? I speak only for myself and my observations. If you cannot follow links I were mentioning thats not my fault. How this is relevant to tagging guideline at all? 

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Mail Threading / was: Wiki Edit War ...

Tom Pfeifer
In reply to this post by jgpacker
jgpacker wrote on 2015-01-21 15:56:
> I agree.
> Sorry, I thought the previous messages we renecessary so the server could find out the who answered to who and to which thread this belongs.

No that is what References (previous Message-IDs in the thread)
in the header are for, in your mail:

Message-ID: <CAE85UivnzFpmkoaFo8+nDtVertaz4UBF=[hidden email]>
In-Reply-To: <[hidden email]>
References: <[hidden email]>
  <[hidden email]>
  <CAJKJX-SXObknsnS5Td0_vh0m=-[hidden email]>
  <CAOnY5KEMpXVXDXQwNTcDgUDENW9t=[hidden email]>
  <[hidden email]>
  <[hidden email]>
  <CAE85UisuePS2XNwZs0oMy5-saGysdqM6+eHyzF=[hidden email]>
  <[hidden email]>
  <[hidden email]>
  <[hidden email]>

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Mail Threading / was: Wiki Edit War ...

jgpacker

Hi,
Richard managed to explain to me in a private email after that email, but thanks for the intention!

jgpacker wrote on 2015-01-21 15:56:
> I agree.
> Sorry, I thought the previous messages we renecessary so the server could find out the who answered to who and to which thread this belongs.

No that is what References (previous Message-IDs in the thread)
in the header are for, in your mail:

Message-ID: <CAE85UivnzFpmkoaFo8+nDtVertaz4UBF=[hidden email]>
In-Reply-To: <[hidden email]>
References: <[hidden email]>
  <[hidden email]>
  <CAJKJX-SXObknsnS5Td0_vh0m=-[hidden email]>
  <CAOnY5KEMpXVXDXQwNTcDgUDENW9t=[hidden email]>
  <[hidden email]>
  <[hidden email]>
  <CAE85UisuePS2XNwZs0oMy5-saGysdqM6+eHyzF=[hidden email]>
  <[hidden email]>
  <[hidden email]>
  <[hidden email]>

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging



If you reply to this email, your message will be added to the discussion below:
http://gis.19327.n5.nabble.com/Wiki-Edit-War-on-using-avoiding-semicolon-lists-tp5830523p5830884.html
To unsubscribe from Wiki Edit War on using/avoiding semicolon lists, click here.
NAML
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

althio
In reply to this post by moltonel 3x Combo

First point:
It is good that several people invent, propose and use various schemes.

Second point:
At some point, unification of schemes with similar intent would be great.
As usage grows, having the same kind of data treated differently is a pain for everyone, mappers, developers, maintainers and data consumers alike.
I don't think one of the schemes is clearly superior to the other, only I wish that it could be open to discussion, open to improvements and settled. Once it is agreed upon (or enforced by any committee, anyone?), people can start to work on unified tagging, clear documentation, adapted presets and simpler algorithms with less cases or exceptions to handle.

Or it is decided that we continue with the two schemes, that both are valid, accepted, documented for tagging and consumption.


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Jo-2
Which too schemes? I think you'd need to be more specific.

As for deprecating semicolon-delimited lists. I don't think it's practical to abolish them completely, so they'll have to stay.

I do agree that it makes sense to try and avoid them as much as possible, but it's simply not always possible.

As for the remark that route_ref shouldn't exist, and that information should be in route relations. Well there are bus lines which may never have route relations. (on demand buses which don't have a fixed route, school buses, market day buses, Sunday and Friday evening special service fares to get students home, etc.). But it's still information which is not hard to map when surveying, and which is good to have when no route relation has been set up yet, or to validate the correctness of the route relation.

About the remark that it's hard to figure out whether an item is missing from the list. I'm sorry, but it's not because 7;8;10;11 are there that 9 necessarily is missing from the list.


I deleted everything which was hidden under the three dots.

Jo

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

althio
Hi Jo/Polyglot and list,

On 22 January 2015 at 12:01, Jo <[hidden email]> wrote:
> Which too schemes? I think you'd need to be more specific.

1. key=values;separated;by;semicolon
2. several key:subkey=*

> route_ref:De_Lijn=1;2;3;4;5;6;7;8;9;284;285;310;315;316;317;333;334;335;337;351;352;358;370;371;372;373;374;380;395;520;524;525;537;586;597

I don't know if this is a real or fictitious example, but try not to
hit the limit of 255 characters for values. ;)

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Jo-2
121 characters... there's still some breathing room. I guess the risk of the street getting congested is higher than hitting the 255 characters limit.

Jo

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

althio
In reply to this post by Friedrich Volkmann

> I even find the second example more difficult to visualize. It's just worse
> than the first in every respect.

From a mapper's point of view
My little +1 for key:subkey=*

In free text like this thread, several key:subkey=* may look more heavy and complicated than key=values;separated;by;semicolon.
_However_ I think this is reversed in the context of editors (iD, JOSM...) and elements lookup [1] where key and values are presented in tables.
+ key:subkey=* tabulated is easier to read
+ key:subkey=* tags are separated, it is slightly easier to select them and update, to delete one only, to add by copy/paste.
+ key=values;separated;by;semicolon means less typing/keystrokes but this is much mitigated by use of presets, auto-completion or copy/paste.


[1] http://www.openstreetmap.org/way/106464005


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Marc Gemis

On Thu, Jan 22, 2015 at 12:59 PM, althio althio <[hidden email]> wrote:

> I even find the second example more difficult to visualize. It's just worse
> than the first in every respect.

From a mapper's point of view
My little +1 for key:subkey=*


New people don't know how to add new keys. So they will have problems to add cuisine:antwerp = yes (in case such a thing would exist :-) ).
it's much easier to come up with cuisine=Antwerp, especially when there is an input field "cuisine".

regards

m.


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Jo-2
Rattling on about those bus stops, I have an other example:

https://www.openstreetmap.org/node/485938967

bus yes
highway bus_stop
name Porte de Hal - Hallepoort
name:fr Porte de Hal
name:nl Hallepoort
network DLVB;IBXL;TECB;TECC;IBXL
operator De Lijn;STIB/MIVB;TEC;STIB/MIVB
public_transport platform
ref:De_Lijn 301026
ref:TECB Bsgipha1
ref:TECC Cbxhal1
route_ref 27
route_ref:De_Lijn 136;137
route_ref:TECB W;123
route_ref:TECC 365a
zone:De_Lijn 20
zone:TEC 6220

A bus stop served by 3 operators, of which one, there are 2 entitities each assigning their own identifier.
For operator and network there are ; in those lists and I don't see what's the problem with those. For ref I don't want to find multivalue. For the rare occasions where this  occurs (2 stops with different refs from the same operator), the nodes are duplicated, then grouped together in a stop_area.

And before anybody says: we don't want those foreign keys in OSM, well the scripts I'm developing to assist contributors for adding route relations, heavily depend on them.

Polyglot

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

dkiselev
In reply to this post by Никита
As I think,

replace
    key=val1;val2
with
    key:val1=*
    key:val2=*

Is just a transposition of problem.

Yes, it's easier now to test for any particular value,
but harder to get all the values inside category.

For key=val1;val2 you need some kind of regex
or smarter application which allows you to find particular value inside collection.

But it's harder to get all tags in category.
How would you get all the payment methods, not the exact 'ellectrico'?

/payment:(.*?).*/
reg. expressions, again?


As for for me as a programmer, both of these approaches are the same.

But for mappers, who are normal people, not the crazy data miners

route_ref=1;2;3;4;5;123;124;125;126;234;235;236;456;457;458a

is easier than

route_ref:1=yes
route_ref:2=yes
route_ref:3=yes
route_ref:4=yes
route_ref:5=yes
route_ref:123=yes
route_ref:124=yes
route_ref:125=yes
route_ref:126=yes
route_ref:234=yes
route_ref:235=yes
route_ref:236=yes
route_ref:456=yes
route_ref:457=yes
route_ref:458a=yes

Anyway none of programmers couldn't be freed out of burden to support both of them.
But, at least we could try to establish one delimiter.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Martin Vonwald (Imagic)

2015-01-22 15:48 GMT+01:00 Dmitry Kiselev <[hidden email]>:
Anyway none of programmers couldn't be freed out of burden to support both of them.
But, at least we could try to establish one delimiter.

Thanks for those clear and true words.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

althio
In reply to this post by Marc Gemis
>> My little +1 for key:subkey=*
>
> New people don't know how to add new keys. So they will have problems to add
> cuisine:antwerp = yes (in case such a thing would exist :-) ).
> it's much easier to come up with cuisine=Antwerp, especially when there is
> an input field "cuisine".

New people can have problems or make mistakes and then experienced
users can help and point to recommended tagging or explain good
practices .
Or do you refer to iD (as the main editor for new people) where it is
not possible to override presets to edit keys on the first part of the
tag panel?

In this thread we are also most interested in multiple values. Not
only cuisine=unique_cuisine. Even for a cuisine as unique as Antwerp
cuisine. :-)
And new people don't know how to add multiple values (well, they know
nothing, don't they?).
In the spirit of free tagging in OSM their first input of multiple
values might be:
"cuisine=antwerp and belgium"
"cuisine=Antwerp chips"
"cuisine=antwerp+fish"
"cuisine=Antwerp_waffles"
and with the automatic completion of iD for new keys (under 'all tags'
in the second part of the panel) finally:
"cuisine=antwerp"  +  "cuisine_1=mussels"

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Wiki Edit War on using/avoiding semicolon lists

Marc Gemis

On Thu, Jan 22, 2015 at 4:49 PM, althio <[hidden email]> wrote:

> New people can have problems or make mistakes and then experienced
> users can help and point to recommended tagging or explain good
> practices .

Not everybody reaches out to community for help. Probably many just stop mapping, requiring them to create a new key, instead of typing something in a free text field is not going to help IMHO.

> In this thread we are also most interested in multiple values. 

I know :-)


> Or do you refer to iD (as the main editor for new people) where it is
> not possible to override presets to edit keys on the first part of the
> tag panel?


What I tried to explain is that when you go for a tagging scheme where only
cuisine:xxx=yes is allowed, the editor (iD) should offer a simple UI that allows people to create new "values". In this case that means keys, since the values are actually in the  keys.

At this moment, it is also not possible to create JOSM presets that generates keys based on user input AFIAK.

Using a xxx:yyy schema also requires checkboxes besides every existing value in JOSM presets.
So I don't see how it is any easier for new mappers or preset creators.

regards

m




_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
1234567