hi. maybe this topic could give ideas on alternatives. (I’ve seen that the new platform helps greatly overcome the language barrier :smile:)

p.s.: not only on coastlines … have you ever seen a town in Morocco? it’s only sharp corners, unless when mapped by Apple.

sometimes corners are so sharp, that Apple prefers to go through walls.

https://forum.openstreetmap.org/viewtopic.php?id=72833

that’s about the coastline edits by the #adt team, with a reply from the team leader. most interaction happened on a public Telegram group.

in what is currently my area of concern, the ADT team announced they would review internal waters; then they edited the coastline.

it’s a pity they did not announce editing the coastline, because in the Panamanian Pacific isis a difficult task because of 5m high tides, potentially 800m distance from high and low tide coastline, and because the real coastline is hardly visible under the mangroves. they could have used some local hints.

also here, when editing the coastline, they did as Charlie described: multiplying by a not negligible factor the amount of nodes. take for example this mangroves island:

https://www.openstreetmap.org/way/142835431/history

from 39 nodes to 267, without changing the shape but looking like they performed a spline smoothing on the original shape, then split every so many metres.

what I would call unnecessary bloating of the database.

I would not know about GitHub, I still don’t manage to blame it for anything, but as for HOT, it is surely not doing a good service to local communities here. Again SColchester, take the two projects 12926 and 12927. SColchester has created them for cenepred-peru, setting his name as “author”, to later pass responsibility to Peter Esquen, not recognizably associated to HOT, and not answering calls by local “super mapper” ovruni.

maybe two separate fields: author and owner would make things more transparent?

maybe a review of your onboarding procedure?

saw recently a comment by HOT Voting Member Céline Jacquin, stating that HOT is not an external organization to OSM. I don’t remember her OSM user name and I can’t provide the link to the statement, which was made in a Telegram group.

After an initial bit of action, it does not look like HOT is really taking measures: SColchester now owns 297 projects in Guatemala, 45 in Botswana, 24 in Zambia. 366 globally, not just HOT hosted, but HOT owned.

On SColchester’s name there’s also 45 DRAFT projects in these three countries, plus 9 in Trinidad and Tobago.

As this sounds so diametrally contrary to the declared intention, and as I’ve written in my follow up diary entry, I give up even providing updates of the measurements.

Hi Minh Nguyen!

about San Jose in California, my guess is limited to name:en and name:es, respectively San Jose and San José. IIUC, the city council decided that the official city name should be in Spanish. in this situation it does not surprise me that the value for name is seen as debatable.

about accents, I’ve noticed that Italians have a better ear for them, funny enough: we don’t write them except on the last syllable.

a tree is a node, and a node is a mathematical point, with no area. have a look at the taginfo page for natural=tree, you will see that some trees (1.3% of the total) have a diameter_crown value, which is very likely what you need. but 1.3% will probably not help you much. you might infer the diameter_crown from the species and the height, but that’s probably a huge lot of work. I guess you need to think further on what you want to achieve related to what information you have.

hi ‘badenk’. You have all my sympathy regarding the poor experience with Andy as a representative of the DWG. You do not have my support for your changeset commenting style.

Have a look at your changeset description statistics on neis-one. You consistently use a standard form of commenting, limiting yourself to something like “corrected tags and points ", approximately in 12200 of your 12450 changesets. Well, obviously you corrected tags and points (or alignments) in the changeset extension box. there's little else you can do with a geographic database, other than alter non-geographic and geographic data (tags and points).

You yourself provide a much better changeset comment, as a reply to a motivated local mapper who managed to have you explain your intentions — what you were doing, on what you based your contribution. The problem is that this came after the local mapper expressed doubts about your changeset, and after you judged the local mapper worthy of an answer (if I may so rephrase your initial statement, how you decide whether or not to reply on comments you receive).

If you would anonymously and collectively consider the whole of OSM worthy of the same consideration, and provide a description of your: intentions; motivations; reasoning on which you base a changeset, as a changeset comment, each time, instead of your repetitive “I did what I did”, you would spare OSM a lot of time-waste and friction, and yourself all this unpleasant DWG attention.

oh, and don’t forget to either delete your construction lines, or to clearly tag them as such, if you insist in keeping them into this global database with millions of contributors and thousands of reviewers.

see you out there, Mario.

I guess the goal was to make one lose interest.

HOT, you win.

hola todavía estoy leyendo, y me parece oportuno comentar este segmento a propósito de la licencia GPL de GNU/Linux: »ya que ninguna empresa es dueña del código«.

debemos distinguir entre propiedad y licencia de uso. hay empresas que son dueñas de (secciones de) código GPL. pueden ser hasta propietarias al 100% de un software, y decidir de liberarlo bajo la licencia GPL. lo que pasa con un código GPL es que todo el mundo puede usarlo, leerlo, mejorarlo, volverlo a distribuir en forma alterada, quedando propietario de sus mejoras, pero sin tener la autorización a cambiar la licencia. (a menos de ponerse de acuerdo con todos los demás propietarios del código).

bueno, sorry, es un punto de vista basado en el software, y todavía me falta todo el resto de tu “diary entry” por leer. (uso OSM en inglés).

ciao, gracias, sigo leyendo.

muy bonito, davvero molto carino.

ah, le code source il est en svg. ça c’est tres bien.

il faut le traduire … :-)

evidentemente dialogare grazie alle traduzioni automatiche non aiuta la mutua comprensione. e che ci vuoi fare.

per non appoggiare nessun imperialismo linguistico, e non sapendo scrivere in esperanto (che comprendo solo con molta fatica), commento in una lingua a caso, quella che mi hanno insegnato da bambino a scuola.

nel 1970 in Cile c’era un gruppo che cantava questi versi, spero ti faccia piacere vederli scritti:

… echaremos fuera el yanqui
y su lenguaje siniestro

poi però furono loro a vedersi esclusi dal loro stesso continente.

ciao, que disfrutes del mapeo y suerte!

Hola Mario Nolasco, justamente en estos días se reanudan intentos de juntarse entre latinoamericanos. https://wiki.openstreetmap.org/wiki/LatAm/20220129_Reuni%C3%B3n_virtual

Hi John, thanks for this diary entry,

I interpret your hint as having an extra post-validation quality check, on each HOT project. once the project gets closed, moves from Published to Archived, it would need going through this sort of acceptance step. I have followed projects in Panamá, being closed just like that, and I think that such an acceptance step would be quite useful.

Regarding the polygon of projects, I use this tiny python script to grab the perimeter of HOT tasks in osm format. Once you open it in JOSM, make sure you keep it in a separate layer apart from the OSM data, or you might inadvertently upload it to OSM.

#!/usr/bin/env python3

import sys
import requests

if len(sys.argv) != 2:
    sys.exit(-1)

info = requests.get(f"https://tasking-manager-tm4-production-api.hotosm.org/api/v2/projects/{sys.argv[1]}/queries/aoi/")
print("""\
<?xml version='1.0' encoding='UTF-8'?>
<osm version='0.6' upload='never' generator='JOSM'>""")

i = 0
points = info.json()['coordinates'][0][0]

for point in points:
    i += 1
    print(f"  <node id='-{i}' lat='{point[1]}' lon='{point[0]}' />")

print(f"  <way id='-{len(points)}'>")

for i in range(len(points)):
    print(f"    <nd ref='-{i+1}' />")
print("    <nd ref='-1' />")

print("""\
  </way>
</osm>""")

my interpretation of your statement takes me back almost 30 years in Italian television, a late night show by satirist Paolo Rossi (not the football player) on the national third channel, good for a potential audience of possibly 2% of all Italians owning a TV set.
in his second issue he felt the need to introduce his definition of vulgarity, because all prudish censors from the main stream news accused him of foul language … and he went on with a vivid representation of vulgarity, between form and content, so strong that 30 years afterwards I’m still considering it a perfect description of things to come, which have come.
(that piece of television art has obviously died under censorship, never to be seen again.)
in my opinion Etiquette is fine for a dinner with your boss or in the entourage of the Queen, but imposing it above the content of the discussion is killing the discussion.

in one month from today 2021-10-17 ESRI will be celebrating their GIS Day, reason enough for a new update. my reports from this data were not looking particularly consistent, so I’ve written a small script to improve repeatability, and I need to update the results for the previous dates:

2021-08-16

• archived projects, prior to 2020-01: 4369
• projects still open, prior to 2020-01: 1710
• … of which 100% mapped and validated: 392
• open projects, prior to 2020-01, directly owned by hot: 87
• … of which 100% mapped and validated: 12
• not-any-more-active bgirardot still owns ‘hot’ projects: 28
• … of his projects, 0% validated: 21

2021-09-05

• archived projects, prior to 2020-01: 4376
• projects still open, prior to 2020-01: 1703
• … of which 100% mapped and validated: 390
• open projects, prior to 2020-01, directly owned by hot: 86
• … of which 100% mapped and validated: 11
• not-any-more-active bgirardot still owns ‘hot’ projects: 28
• … of his projects, 0% validated: 21

2021-09-27

• archived projects, prior to 2020-01: 4411
• projects still open, prior to 2020-01: 1661
• … of which 100% mapped and validated: 387
• open projects, prior to 2020-01, directly owned by hot: 86
• … of which 100% mapped and validated: 11
• not-any-more-active bgirardot still owns ‘hot’ projects: 28
• … of his projects, 0% validated: 21
• CourtneyMClark archived/not-archived projects: 115/29

2021-10-17

• archived projects, prior to 2020-01: 4423
• projects still open, prior to 2020-01: 1654
• … of which 100% mapped and validated: 386
• open projects, prior to 2020-01, directly owned by hot: 86
• … of which 100% mapped and validated: 11
• not-any-more-active bgirardot still owns ‘hot’ projects: 28
• … of his projects, 0% validated: 21
• CourtneyMClark archived/not-archived projects: 117/27

for some reason, Rory Nealon has archived “bang-boom” all HOT-hosted YouthMappers projects in Panamá. this does sound like there won’t be any evaluation, and I do hope I’ll be proven wrong. but this is a YouthMappers issue, it is not a HOT problem, or is it?

Hi CourtneyMClark,

Thanks for the follow-up here in this page. My writing to you was definitely not meant as a “request” —I would not dare!— more as you said later, for letting you know.

While it stays open, I will occasionally come back to this sleeping issue and check how it follows. The updated statistics say:

• 1451 open projects prior to 2020-01,
• 400 of which 100% mapped and validated,
• of the 1451, 80 are directly owned by hot,
• of the 400 100% mapped and validated but still open, 7 are directly hot.
• not-any-more-involved bgirardot is still owner of 28 “hot” published projects,
• among his projects, 21 are 0% validated and only #4762 is 100% mapped and validated.