Proposal talk:Bike safety
No. Absolutely not.
Tags in OSM must be objective and verifiable. Proposals like this do not meet this basic criterion.
Please see [1].
If you want to have a bike map that renders roads differently dependent on average speed and traffic levels, please suggest tags for average speed and traffic levels (objective measurements), not an arbitrary scale based on arbitrarily chosen cut-off points.
--Richard 10:36, 28 September 2011 (BST)
- +1 --Derick 10:40, 20 September 2011 (BST)
- +1 --Beelsebob 10:54, 28 September 2011 (BST)
- +1 --EdLoach 11:06, 28 September 2011 (BST)
- +1 --Cobra 11:17, 28 September 2011 (BST)
No, sorry. Subjective judgements do not belong in a cartographic database. Verifiable features (width, surface, hazards, barriers, speed limits) do, and you're free to make your own subjective evaluation based on those, and perhaps produce a map/rendering based upon your ideas of what constitutes "bike safety". I strongly suspect my idea of what makes a "safe" bicycle route differ significantly from yours. --Norky 10:50, 28 September 2011 (BST)
- If all the statements above are widely accepted, why the hell are we not deleting the 'smoothness' tag ? --Pieren 13:54, 28 September 2011 (BST)
- When given two roads to compare, people will usually agree on which road is smoother. This is not true for bike safety/friendliness/etc. --NE2 17:41, 29 September 2011 (BST)
- If all the statements above are widely accepted, why the hell are we not deleting the 'smoothness' tag ? --Pieren 13:54, 28 September 2011 (BST)
Do you really think that it would be workable to tag average speed, traffic levels, highway width etc, and have one rendering algorithm able to process all this to help a cyclist choosing its way? And what about additional safety/hazard criteria that we would then have to add (frequency of door opening of parked cars, average number of parked cars per km, average number of entry crossing per km...) and then modify the complex rendering process? Decades would pass before something usable would be ready. I can understand that not all cyclist demand the same things to feel safe. That's why the proposed tag is not a binary safe/unsafe value. Its a scale, and after that each cyclist can choose what levels they prefer to ride. --ger4rd 21:29, 28 September 2011 (BST)
- None of the previous attempts at a bike safety/enjoyability scale (found on this wiki) have not given mappers any concrete criteria of what makes a the route better or safer than the next street. There's just mentions of traffic density and potholes, but that doesn't give anyone advice which value they should choose. If you can give such recommendations and demonstrate that it results in better routing in your neighbourhood (after you've tagged them), maybe other's will pick up the tags as "easier to consolidate all these conditions to one value" rathen than start tagging them separately. Which is less expressive data, though, and if you had the guidelines, it would be feasible to make a preprocessor that computes a safety scale for each way, with the given data about traffic density and-what-you-have. Maybe there's a cycling safety study somewhere that has scientific (oh the statistics) knowledge of what makes a route safe or unsafe. Alv 09:15, 3 October 2011 (BST)
No way. Safety is not a linear concept reducible to a single number. The safest place for my toddler might be a local bike path, but that's a very unsafe place to send a road racer at 25mph. Describe the road, indicate any local cycle network routes, and let algorithms do the rest. Brycenesbitt