- From: Melvin Carvalho <melvincarvalho@gmail.com>
- Date: Fri, 24 Mar 2023 18:40:58 +0100
- To: Benjamin Goering <ben@bengo.co>
- Cc: Ryan Barrett <public@ryanb.org>, Bob Wyman <bob@wyman.us>, aaronngray@gmail.com, public-swicg@w3.org
- Message-ID: <CAKaEYhKw1DPu03z6wSrT8Ow=e5hhhsYBf=izzS-8X_84oukSvg@mail.gmail.com>
pá 24. 3. 2023 v 18:34 odesílatel Melvin Carvalho <melvincarvalho@gmail.com>
napsal:
>
>
> pá 24. 3. 2023 v 18:23 odesílatel Benjamin Goering <ben@bengo.co> napsal:
>
>> http content negotiation is normatively required by the activitypub spec
>> (3.2).
>> I'll also point out that some URIs from the TRs already support it.
>>
>
> How do you ensure that each accept header gives back the same triple?
>
> For example, do they in this case.
>
> Not a pedantic point, I've been trying to fix this for years in Solid, and
> its made harder by the fact that the developers dont know how it works. In
> fact the maintainer of the reference implementation said to me recently:
> "What's content negotiation". It's hard to fix these things.
>
Let me add an extra point
Pretty much all the conneg ive seen is broken (e.g. on mastodon too)
It's generally broken on vocabs ive seen too. But nobody actually cares.
Because, people dont actually use vocabs. They may in future e.g. for
xsd:typing but they are barely noticeable today. The whole overhead of the
thing I would argue provides very little developer value other than a
documentation where they can copy and paste.
>
>
>>
>> ```
>>
>>> bengo@bengo ~ ⚡ curl -H "Accept: application/json"
>>> https://www.w3.org/ns/activitystreams | head -n 10
>>> % Total % Received % Xferd Average Speed Time Time Time
>>> Current
>>> Dload Upload Total Spent Left
>>> Speed
>>> 100 7985 100 7985 0 0 39142 0 --:--:-- --:--:-- --:--:--
>>> 40125
>>> {
>>> "@context": {
>>> "@vocab": "_:",
>>> "xsd": "http://www.w3.org/2001/XMLSchema#",
>>> "as": "https://www.w3.org/ns/activitystreams#",
>>> "ldp": "http://www.w3.org/ns/ldp#",
>>> "vcard": "http://www.w3.org/2006/vcard/ns#",
>>> "id": "@id",
>>> "type": "@type",
>>> "Accept": "as:Accept",
>>> bengo@bengo ~ ⚡ curl -H "Accept: text/html"
>>> https://www.w3.org/ns/activitystreams | head -n 10
>>> % Total % Received % Xferd Average Speed Time Time Time
>>> Current
>>> Dload Upload Total Spent Left
>>> Speed
>>> 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:--
>>> 0<!DOCTYPE html><html dir="ltr" typeof="bibo:Document " prefix="bibo:
>>> http://purl.org/ontology/bibo/ w3p: http://www.w3.org/2001/02pd/rec54#
>>> as: https://www.w3.org/ns/activitystreams#" lang="en"><head><meta
>>> charset="utf-8"><meta name="generator" content="ReSpec 20.7.1"><meta
>>> name="viewport" content="width=device-width, initial-scale=1,
>>> shrink-to-fit=no"><meta property="dc:language" content="en" lang="">
>>> <title>ActivityStreams 2.0 Terms</title>
>>>
>>>
>>>
>>>
>>> <style>/*****************************************************************
>>> * ReSpec 3 CSS
>>> * Robin Berjon - http://berjon.com/
>>> *****************************************************************/
>>>
>>> 100 33068 0 33068 0 0 148k 0 --:--:-- --:--:-- --:--:--
>>> 153k
>>> curl: (23) Failure writing output to destination
>>>
>> ```
>>
>> On Fri, Mar 24, 2023 at 10:16 AM Ryan Barrett <public@ryanb.org> wrote:
>>
>>> Sure! And feel free to redirect to a different venue if we're getting
>>> off topic. Off the top of my head:
>>>
>>> 1. Most web developers are not aware that conneg exists. This is
>>> usually fine, since it's uncommon, but can be an unpleasant surprise when
>>> they first hit it and don't understand it.
>>> 2. The `Accept` header is notoriously complicated to parse,
>>> generate, and fully comply with. Imo much of that complexity is unneeded in
>>> practice.
>>> 3. It often breaks caching when it's first introduced. Internal and
>>> external caches generally ignore Content-Type and Accept (etc) by default,
>>> so when conneg is first introduced, clients often get the wrong response
>>> type. Developers learn the hard way that they need to add Accept to Vary
>>> and to their framework's cache.
>>> 4. URLs leak across contexts. They may start out within a single
>>> service or API, but they gradually proliferate into user-visible links,
>>> object identifiers in other systems, add-on services, third party tools,
>>> etc. These generally don't know the original conneg requirements, which
>>> causes unpleasant surprises.
>>>
>>> I think most of this boils down to: modality generally considered
>>> harmful. When something always behaves the same way, it's reliable and easy
>>> to use. When it behaves differently based on something far away in the
>>> environment that you may not know exists, it's unreliable and surprising.
>>> Add in a very large ecosystem of independent tools that all need to
>>> interoperate, often in fine-grained ways, and you have a recipe for failure.
>>>
>>> On Fri, Mar 24, 2023 at 9:42 AM Bob Wyman <bob@wyman.us> wrote:
>>>
>>>> Melvin, Ryan,
>>>> Could you provide at least a short summary of why "content negotiation"
>>>> has not lived up to expectations? It is good to know that it is
>>>> non-optimal. It would be better to know why.
>>>>
>>>> bob wyman
>>>>
>>>>
>>>
>>> --
>>> https://snarfed.org/
>>>
>>
Received on Friday, 24 March 2023 17:41:25 UTC