What’s Ruby local/official/test?

大したものですね、どうも！

@MxxCon I think you’re missing a couple of points.

The (possible) problem is not, using non-open communication channels. It is, not having the relevant information (whatever it might be, e.g. financials in this whole drama) in open communication channels.

If said information is available on GitHub (as you said), which is open for viewing even without an account, then I’d say that’s fine (unfortunate, IMO, a better option would be e.g. the Wiki, but fine nonetheless!).

If said information is only available in a Slack/Discord/whatever group (which I believe is not open for viewing without an account), then that is a problem due to the “Commitment to Open Communication Channels”

BTW note that the “Commitment to Open Communication Channels” applies only to OSMF communication channels, but not to local communities that manage/administer their own communication means (not sure about “local chapters” though?).

@MxxCon

Anybody is free to create an account there. There’s no requirement that OSM must make broadcast announcements in your local free newspaper for you to see about it.

Being free is different from being open. And in case you’re not aware, take a look at https://wiki.osmfoundation.org/wiki/Commitment_to_Open_Communication_Channels

First of all, I’m sorry to hear a contributor as active as you has decided to leave. It is always a shame and a big loss. OSM has come this far because we’re many. There is a saying in Portuguese: “grão a grão enche a galinha o papo” (lit. “grain by grain fills the chicken its belly”).

As others have also already said, the AWS spenditure picture thread is quite different from what’s painted in this diary. I haven’t followed up on the other points because my time is limited and I’d rather spend it on getting shit done.

there are core principles I stand by that deeply inform my perspective.

Core principles may be informed by facts, observations, etc. But core principles cannot inform perspectives, only influence.

One of my core beliefs is the importance of creating and utilizing software that respects its users. I have a profound aversion to the practices of many big-tech companies. In my eyes, they commodify human beings and often operate in ways I find ethically objectionable.

I share with you this principle/belief. I thank you for striving and putting in effort for a better world in this front, such as by contributing to OSM and FLOSS, and I hope you continue to do so for many years to come!

However, at the heart of an open project (such as OSM), we must put emphasis on us, the humans, and our interactions. If not for codified norms (CoC, ethical, law, cultural/social norms, …), at least for the prosperity of the said open projects. Because humans are not a commodity, as you so said, and without the human contributors the said open projects would not exist. So try to give your fellow humans reasons to continue in the project. If, no matter why, you can’t give reasons to continue, try at least not to give them reasons to leave.

Unfortunately, from several of your posts I saw more reasons not to engage with you than reasons to engage with you, and I can only imagine others felt the same. I just hope none of the core pillars of this project (such as @Firefishy of the OWG) have felt it in them reason to leave.

I also recommend Organic Maps and OsmAnd – the latter for its complexity, the former for its simplicity.

Edits made with Organic Maps while offline will be uploaded once you go online.

@2hu4u I see. I was only commenting on your appeal to cooperate because for that the wiki would very much be a better place (better discoverability, anyone can contribute, and no need to go through Google).

Wouldn’t the wiki be a better place? There are some pages comparing hardware used for OSM things already.

Thanks for your contributions!

Just a bit of food for thought: when I started I thought POIs were more important than, for example, buildings. My reasoning was that, in general, people are more likely to be looking for a POI than for a building. But my opinion has changed meanwhile, because buildings, especially if they have address tags, are very useful as reference points for mappers, if you’re not certain of the location of a POI, for example.

So if one day you feel tired of adding POIs try adding some buildings instead, they might help you in the future :)

Totally agree that calling tab-delimited CSV “CSV” is silly!

About the human-readable part: that could be a reason, but I don’t think that’s it… There’s so much that can be said about “human-readable-ness” I don’t even know where to begin, apart from the fact that “human-readable” is ambiguous.

In the sense of being (or not) “text”: both less and vim display these characters, so they’re definitely readable. Example file of two records, one of values a1, a2, a3, and another of values b1, b2, b3:

$ xxd test.dsv
00000000: 6131 1f61 321f 6133 1e62 311f 6232 1f1e  a1.a2.a3.b1.b2..
00000010: 1c0a

In the sense of being easy(-ish) to read: I wouldn’t call CSV files that look like this readable:

foo,"bar,baz,zaz",,,,,asdf

How many columns does this record have? How many (non-)empty columns does this record have? What’s the value of the 5th column? You can’t answer any of these questions at a glance, so this CSV file isn’t significantly more readable than the file above.

So if your only reason to use CSV (w/ quoting+escaping) is that it’s human-readable, just don’t. They’re not anyway, they’re meant to be consumed/processed by a program, and for that there are better alternatives.

Finally, we use non-human-readable files all the time (images, audio, video, zips, etc, etc)! I’ve never seen anyone complain their images/audio/etc aren’t human-readable.

@jleedev

No tabular data is “simple”.

That should be obviously false to anyone…

Can you say for sure that nobody has included (…) anywhere in the Host: header of a tile request, ever?

There are two sides to that coin:

1. Of course you can’t!
2. Of course you can!

RE (1): obviously you can’t because it’s user input.

RE (2): like @mmd said, “I’m missing the bigger picture here.” I don’t know what the purpose of these files is, I don’t know who these files are for, I don’t know if they’ll be automatically updated regularly, I don’t know if the files will be published raw (untreated) or not, etc, etc – because these details aren’t explicitly stated anywhere.

For example, if after the raw data is gathered you check that there’s no occurrence of a {tab,space,US,comma,…}, then you can safely use {tab,space,US,comma,…} as the separator of the published file – but this assumes that the data isn’t automatically updated regularly.

Python’s csv.writer with unix_dialect and QUOTE_NONNUMERIC

That’s an implementation detail a user shouldn’t have to think about, and, personally, I have no idea what the unix_dialect is… Additionally, it’s not guaranteed to be compatible across different programming languages/DSV libraries/programs, even though these quoting+escaping rules may look standard. Before I can use the file, I (as the user) have to make sure whatever program I want to use correctly supports all these implementation details.

OTOH, @Jochen Topf suggested using a SQLite file, which looks to me like a much better alternative even if it sounds overkill at first sight. Among many other pros over some DSV format:

• there’s only one SQLite implementation;
• it’s stable as a rock;
• pretty much every language has bindings to it;
• there’s the sqlite CLI program;
• you get the power of the SQL language;
• if you don’t like SQL you can still export to whatever DSV dialect (and therefore use *nix tools);
• there are no choices to make;
• it’s guaranteed to work for everyone everywhere always!

Hopefully that makes clear what I was trying to express: if you can’t guarantee interoperability through simplicity (i.e. without hacks), then at least guarantee interoperability through actual standards.

PS: using tabs instead of comma+quoting+escaping is the opposite of weird, and in general (for what people normally use CSV) it’s a step in the right direction. Python’s CSV module, for example, (since it’s been mentioned) supports “arbitrary” separators as an instance parameter, IIRC.

I fail to understand how this question of which separator to use in a DSV file for simple tabular data still comes up, since it’s been solved for decades, without using subpar “solutions” (e.g. quoting & escaping) that unnecessarily preclude handling the files with simple tools (e.g. *nix shell and related tools (cut, sed, sort, read, etc); a simple string split in a programming language; etc)…

Hi,

What do you mean by “changing my location”? Where do you see that change?

And what did you do to report whoever did that?

I use “you” in the text but I’m not referring to you or any specific person, it’s only to make reading and writing easier (and try to force the reader into thinking from some perspective).

I had never thought about this before. In my opinion it all depends on the person’s circumstances and personal preference/ideology, and the “ground truth”.

Hypothetical scenario 1: your house is in the middle of nowhere, in a very nice secluded place, and you just want some peace and quiet. But because of the data on OSM (specifically, your house and access to it) some hikers show up at your door. This could be for a variety of different reasons, intentional or not, but unless the encouter is positive (maybe they need some help; maybe they’re nice people and you all end up making BBQ together), then it is negative for you. With the data on OSM the encouters are arguably more likely to happen. All that’s left to discuss is if your glass is half empty or half full.

Hypothetical scenario 2: your house is in a city, or even a village will do. There are several houses close to yours, and they’re all mapped. Whether or not your own house is mapped makes no significant difference to whether randos show up at your door – random strangers passing by your door is likely to be the reality already.

My house happened to be mapped already when I started mapping and removing it never even crossed my mind. But my case is closer to (2) above. Should it be more like (1) I would very likely think about it.

Do you have the right to complain or request that your property or its details be removed? Yeah, I think so, why not?

Do you have the right/should you remove your property or some of its details from OSM? Maybe, but probably not. Contacting the “OSM authorities” (DWG?) and explaing your situation is probably a better approach.

Should the “OSM authorities” remove your property or some of its details if you ask them to? I think it all hinges on the “ground truth”. If what you want to remove is publicly verifiable, that is, by anyone with no special access, then I’d say no. In this case removing it from OSM won’t bring you much benefit either, as the data is available elsewhere or can be aggregated elsewhere. And if you do remove it from OSM, it should be considered vandalism. OTOH if the data is sensitive, private, or special access is required to verify it, then I’d say yes – and the relevant history too if possible!

Should your property (or some of its details) be removed if it was added to OSM before it became your property? I don’t know about this one… I’m leaning to no, but I don’t have a strong opinion.

Some examples (IMO):

• An uncovered open-air pool is not private if it can be seen from the street or aerial imgery.
• An indoor pool is private if it cannot be seen from the street.
• The “blueprint” of a house (levels, rooms, stairs, &c) is most likely private.
• Your house in the middle of nowhere is not private if it can be seen from aerial imagery or if one can stumble upon it passing by.

Was it unethical for this user to access someone else’s computer (OpenStreetMap servers) after they had been banned from doing so for the sake of protecting their privacy in an extralegal manner?

I’d reformulate this question a bit first to make it more general, starting by replacing “was” with “is”, and “this user” with “a user”.

I don’t care much about the “extralegal” part – I don’t even know what it means, really. IMO, if someone’s privacy is at risk then they have the right to protect it, even if it means breaking some site’s rules (OSM in this case), as long as in the process they don’t cause any harm. If their privacy is not at ristk, the removal can be considered vandalism, as was this case.

And there’s the technical part that I’m ignoring: anyone could download the history to retrieve their house even after it being removed, because OSM does keep a history. So simply removing the features doesn’t solve their problem. But maybe they weren’t aware of that?

What could I have done better?

Adding the house to WikiData was a dick move, especially since it was out of spite, as you say. You didn’t know their reasons and how it affected them. It very likely made them angrier and more frustrated with OSM, growing ill will between both parties. Now they can go around telling people we’re just some jerks playing cartographers and invading people’s privacy, and with some reason to it.

And I didn’t understand what you meant with “but it is a good act, or at least a neutral one”.

Is it acceptable for the Data Working Group to move straight to a long block just because other users attempted to explain the site’s policies to them first?

“It’s no use trying to speak to someone who doesn’t wish to listen”, or something like that…

It seems several people tried to reason with them, several times, to no avail. Better to cut the losses short and just end the matter.

What do we say to other users with a similar concern? Is being upfront about the legal problems threatening, or merely truthful and effective?

What are these “legal problems”?

I say that we should try to go case-by-case as much as possible. It’s all good to try and come up with some clear-cut rules that say “this is fine, that is not”, but rules don’t always work (and there are priorities, e.g. to me, privacy > rules).

And it must be made very clear that we value people’s privacy and take it very seriously, within reason – if they don’t care to hear us, we needn’t care to hear them.

Welcome and thank you for your contribution!

I would also add that making smaller changesets is good practice. It makes it easier for people to review them, it prevents you from possibly losing some work, and makes it easier to revert changes if needed.

If you have Linux installed on your Chromebook you just have to install Java and download JOSM, ant it should Just Work. Otherwise I don’t know, but I second @GOwin’s suggestion.

Hi there o/

What do you mean by “shift”, and why do you keep saying “shift”?

Very nice, thanks for sharing! The trick to underlay a picture of the map of the mall is smart. How do you do that? I know how to use aerial imagery but didn’t know you could use just any image.

Also, looking at the screenshots, it seemed like the “Pizza Pai” parking was missing a trema in “Pai”?