[Tagging] Feature Proposal - RFC - Discouraging the use of deprecated schemes
Niels Elgaard Larsen
elgaard at agol.dk
Wed Mar 24 23:47:22 UTC 2021
> My opinions follow. While I don't dissuade others from using Slack, Discord and other proprietary comm tools, (and so what if I did? — I'm simply one guy in California with strong opinions...)\
This guy in Denmark feels the same way.
Besides, I have had to use Slack professionally, and it is just horrible. I am not
going back to Slack for my unpaid OSM activity.
> and I realize that a certain amount of sometimes important discussion takes place within them, I personally don't (can't) use them because of their onerous licensing terms: their bar-of-entry is too high. As I have said, I believe that open-source, "least common denominator" tools (like email, this list-server and low-bar-of-entry, "open" and home-grown tools like OSM help forum) are highly preferable in our project, Open being our first name. I know I'm not the only one who feels this way, it is similar to those who disdain big-tech social media, either having quit them or never signing up in the first place. I dream of fully open-source similar tools; it may be that one day we have such things which do not require a draconian, "give it all away" -style license (like Slack, Twitter, Discord, Facebook...). Only glancing at Loomio, I find it promising, there likely are or will be other contenders. Let's keep an eye on Loomio; thank you Seth for mentioning it. Sorry, Sören, Slack is NOT ideal, it is proprietary with onerous licensing terms. Email (and the way it integrates into list-servers) DOES have the advantage of "everybody already knows it" and hence, the desired low-bar-of-entry.
> Despite some downsides, this list-server is one of the few "working" places where (tag) discussions have a lot of the upsides we want. But list-servers can be tedious, requiring a great deal of reading, therefore focused time which is often spent for naught. We should be able to do better, perhaps initially by better fragmenting threads or even whole lists.
> That said, I also agree about "chat-like platforms" (thanks to Paul and others from whom I borrow / repeat):
> • they can get easily side-tracked into a twisty maze of rabbit holes, even with no conflict,
> • they don't scale well, making it hard to encourage more people to participate (and we DO wish to do so),
> • they can be too difficult to follow if there are more than several like-minded people who start out in close agreement,
> • they are synchronous, requiring the difficulty of simultaneous, live participation (rather than asynchronous / non-linear),
> • for serious content discussions they tend to blur issues and have a high risk of conflict rather than cooperation (questioned or somewhat disputed).
> So, while chat-like platforms have their place (a bit hard to describe, I agree), promoting these likely leads to more fragmentation. Consensus begins to emerge that chat-based isn't quite appropriate for this, "forum-based" can be included (have their place), as they are asynchronous, too, and list-servers (like this) work (not ideally, but they work). I am not the only one who doesn't want the "real dread" of mail-lists to stifle participation, yet I agree that's what we have.
> One of the most important things said in this thread (imo) is Martin K.'s "our tagging is consistent, it's the way addresses are associated in the real world that is different." While this is true is about addresses, it may be generalizable to other aspects of OSM data vs. real world data; we should look out for these and be aware of them when we find them. This is another example of "regional-based tagging" being different in OSM because real-world data are different (like highway classification tagging differences). These are an important, identifiable aspect of how tagging does and will diverge from documented form, and it is critical that we continue to stress that they are necessary and allowable, despite their initially apparently odious break from our norms. Other examples include rail tagging differences, rather well done in ORM and its extensive wiki documentation and on a country-by-country basis where necessary: when these happen, we can and do document such differences in an explainable and allowable way, and that does a decent job of making them OK.
> Tagging mailing list
> Tagging at openstreetmap.org
Niels Elgaard Larsen
More information about the Tagging