openstack-doc-core team mailing list archive
-
openstack-doc-core team
-
Mailing list archive
-
Message #00293
Re: Ironic docs & Service name standards
On Thu, Jun 25, 2015 at 10:13 AM, Andreas Jaeger <aj@xxxxxxxx> wrote:
> On 06/25/2015 04:10 PM, Anne Gentle wrote:
>
>>
>>
>> On Thu, Jun 25, 2015 at 5:17 AM, Andreas Jaeger <aj@xxxxxxxx
>> <mailto:aj@xxxxxxxx>> wrote:
>>
>> On 06/25/2015 12:04 PM, Steve Gordon wrote:
>>
>> ----- Original Message -----
>>
>> From: "Andreas Jaeger" <aj@xxxxxxxx <mailto:aj@xxxxxxxx>>
>> To: "Lana Brindley" <openstack@xxxxxxxxxxxxxxxx
>> <mailto:openstack@xxxxxxxxxxxxxxxx>>,
>> openstack-doc-core@xxxxxxxxxxxxxxxxxxx
>> <mailto:openstack-doc-core@xxxxxxxxxxxxxxxxxxx>
>>
>>
>> On 06/25/2015 02:16 AM, Lana Brindley wrote:
>>
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> Hi everyone,
>>
>> First of all, I'm sorry that the Ironic docs thing has
>> turned into a bit
>> of a disaster. I had no idea that we were going to end
>> up in a standards
>> argument over this, but I'm still hopeful that we can
>> see a path through
>> this.
>>
>> The problem: A number of patches (mostly created by
>> Shilla, if I'm not
>> mistaken) are now being blocked in the Ironic repo by an
>> Ironic core who
>> doesn't agree with our standards. While they're
>> certainly entitled to
>> question our standards, blocking the patches means that
>> other good work
>> is not getting merged.
>>
>> The solution: I'm not certain. Right now, what I want to
>> do is send
>> email to Devananda (the Ironic PTL) with a list of the
>> patches being
>> blocked, so we can determine if there's a way we can
>> appease the Ironic
>> core team's concerns over standards. Shilla (and others
>> who have a patch
>> out on this), can you please send me a list of the
>> patches currently
>> being blocked?
>>
>> The fallout: We're already arguing about the standard in
>> question on our
>> own list, and the community seems deeply divided. I
>> personally don't
>> care about the capitalisation of service names, and am
>> happy to enforce
>> whatever the community decides. That said, I feel as
>> though the
>> community is unlikely to come to a solid conclusion on
>> this. Do cores
>> have an opinion? If there is a strong tendency amongst
>> this group,
>> perhaps it's easier to just go with that and adjust
>> accordingly. Please
>> feel free to debate this topic on this thread.
>>
>>
>> So, this is "Ironic" vs "ironic"?
>>
>> I think what needs to be worked out with a project is that our
>> conventions are enforced if we work together. Like there are
>> hacking
>> rules for code which a core team reviews, there are
>> documentation rules
>> that are under the Documentation team (btw. going to a
>> reviewed style
>> guide would help with the story;)!
>>
>> On the Ironic vs ironic: While I prefer the lower-case, I
>> see upper-case
>> everywhere.
>>
>> It's the number one change that I request during reviews.
>>
>> If we want to enforce our documentation style everywhere, it
>> might be
>> better to not enforce the capitalization or change the rule.
>>
>>
>> So, I'm willing to change my vote if that is needed to adopt
>> the Doc
>> style everywhere ;)
>>
>> Anne, is the Doc style adoption a TC discussion?
>>
>> What do we need to make our style guide better consumable by
>> other projects?
>>
>> Andreas
>>
>>
>> This is the patch in question:
>>
>> https://review.openstack.org/#/c/194230/1/reference/projects.yaml
>>
>> It's about Bare metal service versus Bare Metal service.
>>
>>
>> Thanks!
>>
>> Ah, that one ;(
>>
>> So, it's "Bare metal service" vs "Bare Metal service" vs "Bare Metal
>> Service".
>>
>> We did a couple of consistency changes there and might need to go -
>> as mentioned in the review - over the complete list to have
>> consistency. And that discussion we need to have and I think Docs
>> should have final say on it - but up for the TC,
>>
>>
>> Yes, it is up to the TC but they certainly want docs team input. I've
>> put my input on the patch itself:
>>
>> ---
>>
>> For context: we are in this situation because of legal names, where some
>> of the original governance allowed projects to use "OpenStack" in the
>> docs as part of their name very early on. So, sometimes the phrasing
>> needed "service" to help with readability. Also, legally we were
>> required to use module due to branding with the OpenStack name.
>>
>> So, the convention is:
>>
>> * Uppercase first letter of first word only.
>> * Do not use OpenStack in the name of the service.
>> * Use module if it is consuming other services (such as heat).
>> * Use service in general as that is mostly what is being added as
>> projects.
>>
>> Now that we have the interop and brand guidelines, the use of OpenStack
>> in a name in the docs is not needed, but we still need to be very
>> careful and considerate in naming consistently.
>>
>> Bare metal fits our first naming rule.
>>
>> Changing these back and forth is not helpful. I will better document the
>> naming conventions in the governance repo to avoid this in the future.
>>
>>
>> ---
>>
>> But here's what I really want to say. To be high-value contributors to
>> all the projects we must avoid the reputation of being nit pickers.
>>
>
> So, your advise for projects like Ironic would be not to send a number of
> patches that change all conventions and don't do anything else?
>
Yes. Because:
- Their contributors should follow OpenStack conventions if they want to be
published to docs.openstack.org.
- We want docs to be self-service, as in, we give you conventions, you
follow them and review to those conventions. If you disagree with them you
politely bring it up on the -docs mailing list but still agree to follow
conventions.
- Higher-value contributions happen with information architecture, tested
docs, and accuracy/speed in reviews and corrections.
>
> And newly written docs that mix with existing ones? Which conventions
> should those use?
>
> Please review my rules above and see how they work for you in the docs,
>> especially for the newest projects that came in the last two weeks. I
>> will write a patch that describes the guidance for the governance repo
>> based on input here.
>>
>
> One example: Object Storage.
>
> And we should probably merge our wiki page with the text below or make it
> clearer why we use "OpenStack Compute" but "Orchestration module for
> OpenStack".
>
> Current list of services in projects.yaml:
> service: Compute
> service: Object Storage
>
should be storage
> service: Image service
> service: Identity
> service: Dashboard
> service: Networking
> service: Block Storage
>
should be storage
> service: Telemetry
> service: Orchestration
> service: Database service
> service: Bare metal service
> service: Common libraries
> service: Deployment
> service: Message service
> service: Data processing service
> service: Key management service
> service: DNS service
> service: Containers service
> service: Shared file systems
> service: Application catalog
> service: command line client
> service: Governance service
> service: Benchmark service
> service: Workflow service
> service: Key-Value Store as a Service
>
Key-value storage
> service: Puppet modules for the OpenStack components
> service: Chef cookbooks for the OpenStack components
> service: Search service
> service: Ansible Playbooks and Roles for the deployment of OpenStack
>
>
>
Revising "service" guidance: Use service when the provided capability is a
noun (Image, Database) and do not add service when the capability is a
capacity (Compute, Bare metal). Would that work?
> Andreas
> --
> Andreas Jaeger aj@{suse.com,opensuse.org} Twitter/Identica: jaegerandi
> SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
> GF: Felix Imendörffer, Jane Smithard, Dilip Upmanyu, Graham Norton,
> HRB 21284 (AG Nürnberg)
> GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126
>
>
Follow ups
References