ReSpec updated

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
13 messages Options
Reply | Threaded
Open this post in threaded view
|

ReSpec updated

Shane McCarron-7
Okay - we have a stable ReSpec pushed that supports the new stylesheets etc. by default, and has a number of other changes.  Sorry for the delay.  Thanks to Marcos for all his late nights and hard work to get this stable again!

--
-Shane
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Michael Cooper-2
There are still changes impacting publication; this version is adding
"It is expected to become a W3C Note." for a Note-track document that
pubrules rejects; the previous version did not add that, and is accepted
by pubrules. This is the one giving me pain right now but there may be
others. The output of the new respec needs checking against pubrules
before it can be considered fixed. Michael

On 10/03/2016 1:36 PM, Shane McCarron wrote:
> Okay - we have a stable ReSpec pushed that supports the new
> stylesheets etc. by default, and has a number of other changes.  Sorry
> for the delay.  Thanks to Marcos for all his late nights and hard work
> to get this stable again!
>
> --
> -Shane


Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Michael Cooper-2
Shane told me to do a force-refresh and that did fix the proximate
problem. Michael

On 10/03/2016 1:53 PM, Michael Cooper wrote:

> There are still changes impacting publication; this version is adding
> "It is expected to become a W3C Note." for a Note-track document that
> pubrules rejects; the previous version did not add that, and is
> accepted by pubrules. This is the one giving me pain right now but
> there may be others. The output of the new respec needs checking
> against pubrules before it can be considered fixed. Michael
>
> On 10/03/2016 1:36 PM, Shane McCarron wrote:
>> Okay - we have a stable ReSpec pushed that supports the new
>> stylesheets etc. by default, and has a number of other changes.  
>> Sorry for the delay.  Thanks to Marcos for all his late nights and
>> hard work to get this stable again!
>>
>> --
>> -Shane
>
>


Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Shane McCarron-7
In reply to this post by Michael Cooper-2
Interesting - 1) I didn't know we were publishing any notes right now, and 2) Happy to rip that out if it shouldn't be there.  URI?

On Thu, Mar 10, 2016 at 12:53 PM, Michael Cooper <[hidden email]> wrote:
There are still changes impacting publication; this version is adding "It is expected to become a W3C Note." for a Note-track document that pubrules rejects; the previous version did not add that, and is accepted by pubrules. This is the one giving me pain right now but there may be others. The output of the new respec needs checking against pubrules before it can be considered fixed. Michael


On 10/03/2016 1:36 PM, Shane McCarron wrote:
Okay - we have a stable ReSpec pushed that supports the new stylesheets etc. by default, and has a number of other changes.  Sorry for the delay.  Thanks to Marcos for all his late nights and hard work to get this stable again!

--
-Shane




--
-Shane
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Tobie Langel-4
In reply to this post by Michael Cooper-2
On Thu, 10 Mar 2016, at 19:53, Michael Cooper wrote:
> There are still changes impacting publication; this version is adding
> "It is expected to become a W3C Note." for a Note-track document that
> pubrules rejects; the previous version did not add that, and is accepted
> by pubrules. This is the one giving me pain right now but there may be
> others. The output of the new respec needs checking against pubrules
> before it can be considered fixed. Michael

I'm sure a pull request fixing this would be most welcome. :)

--tobie

Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Michael Cooper-2
On 10/03/2016 2:06 PM, Tobie Langel wrote:
> On Thu, 10 Mar 2016, at 19:53, Michael Cooper wrote:
>> There are still changes impacting publication; this version is adding
>> "It is expected to become a W3C Note." for a Note-track document that
>> pubrules rejects; the previous version did not add that, and is accepted
>> by pubrules. This is the one giving me pain right now but there may be
>> others. The output of the new respec needs checking against pubrules
>> before it can be considered fixed. Michael
> I'm sure a pull request fixing this would be most welcome. :)
When I'm in publication hell, exacerbated by a recently broken tool,
trying to learn the code in order to submit a pull request is just not
possible. I just don't have time to deal with that. Michael
>
> --tobie
>


Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Shane McCarron-6
Nor should you.  I will dig it out.  I was just looking for a reference to something that exhibited the behavior.

On Thu, Mar 10, 2016 at 1:13 PM, Michael Cooper <[hidden email]> wrote:
On 10/03/2016 2:06 PM, Tobie Langel wrote:
On Thu, 10 Mar 2016, at 19:53, Michael Cooper wrote:
There are still changes impacting publication; this version is adding
"It is expected to become a W3C Note." for a Note-track document that
pubrules rejects; the previous version did not add that, and is accepted
by pubrules. This is the one giving me pain right now but there may be
others. The output of the new respec needs checking against pubrules
before it can be considered fixed. Michael
I'm sure a pull request fixing this would be most welcome. :)
When I'm in publication hell, exacerbated by a recently broken tool, trying to learn the code in order to submit a pull request is just not possible. I just don't have time to deal with that. Michael

--tobie






--
Shane McCarron
Projects Manager, Spec-Ops
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Tobie Langel-4
In reply to this post by Michael Cooper-2
On Thu, 10 Mar 2016, at 20:13, Michael Cooper wrote:

> On 10/03/2016 2:06 PM, Tobie Langel wrote:
> > On Thu, 10 Mar 2016, at 19:53, Michael Cooper wrote:
> >> There are still changes impacting publication; this version is adding
> >> "It is expected to become a W3C Note." for a Note-track document that
> >> pubrules rejects; the previous version did not add that, and is accepted
> >> by pubrules. This is the one giving me pain right now but there may be
> >> others. The output of the new respec needs checking against pubrules
> >> before it can be considered fixed. Michael
> > I'm sure a pull request fixing this would be most welcome. :)
> When I'm in publication hell, exacerbated by a recently broken tool,
> trying to learn the code in order to submit a pull request is just not
> possible. I just don't have time to deal with that.

Remember open-source contributors don't owe you anything and might be as
exacerbated by your expression of entitlement as you might be by their
broken tool. A tool which you are getting for free. :)

--tobie


Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Michael Cooper-2
In reply to this post by Shane McCarron-6
On 10/03/2016 2:17 PM, Shane McCarron wrote:
Nor should you.  I will dig it out.  I was just looking for a reference to something that exhibited the behavior.
I was testing a local document prior to publishing live, but using the absolute reference to Respec. So hard to pass a URI.

But as I mentioned in a part of the thread that crossed with this part, Shane's suggestion at the tail end of a telecon to force-refresh fixed it. I just hadn't expected that that should be necessary. So no need to spend time investigating.

But nonetheless felt need to push back on "just submit a pull request" because that is a highly unrealistic expectation for most users of Respec. I have a couple times tried to learn the Respec code in order to submit pull requests but quickly got lost; there is no hope of taking time to do that when I'm trying to get a publication out under pressure of committed dates.

Michael

On Thu, Mar 10, 2016 at 1:13 PM, Michael Cooper <[hidden email]> wrote:
On 10/03/2016 2:06 PM, Tobie Langel wrote:
On Thu, 10 Mar 2016, at 19:53, Michael Cooper wrote:
There are still changes impacting publication; this version is adding
"It is expected to become a W3C Note." for a Note-track document that
pubrules rejects; the previous version did not add that, and is accepted
by pubrules. This is the one giving me pain right now but there may be
others. The output of the new respec needs checking against pubrules
before it can be considered fixed. Michael
I'm sure a pull request fixing this would be most welcome. :)
When I'm in publication hell, exacerbated by a recently broken tool, trying to learn the code in order to submit a pull request is just not possible. I just don't have time to deal with that. Michael

--tobie






--
Shane McCarron
Projects Manager, Spec-Ops

Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Tobie Langel-4
On Thu, 10 Mar 2016, at 20:26, Michael Cooper wrote:
On 10/03/2016 2:17 PM, Shane McCarron wrote:
 
Nor should you.  I will dig it out.  I was just looking for a reference to something that exhibited the behavior.
I was testing a local document prior to publishing live, but using the absolute reference to Respec. So hard to pass a URI.
 
But as I mentioned in a part of the thread that crossed with this part, Shane's suggestion at the tail end of a telecon to force-refresh fixed it. I just hadn't expected that that should be necessary. So no need to spend time investigating.
 
But nonetheless felt need to push back on "just submit a pull request" because that is a highly unrealistic expectation for most users of Respec. I have a couple times tried to learn the Respec code in order to submit pull requests but quickly got lost; there is no hope of taking time to do that when I'm trying to get a publication out under pressure of committed dates. 

This feeling of entitlement to open-source software is why open-source developers burn-out and end up abandoning development of their projects altogether.

So what might seem a highly unrealistic expectation today, might be something you'll have to figure out on your own in the not so distant future.

That said, I don't have any insider information on Marcos and Shane's plans. They might still be around in a decade working on Respec for all I know.
 
--tobie
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Michael Cooper-2
In reply to this post by Tobie Langel-4
I'm sorry, I don't mean to sound entitled, but recognize that the wording of my message does sound that way. I know Respec is a volunteer project made available for free. I would like to be able to contribute and share the load, but it uses a technology I just don't have skills in. So I have to depend on the people who do maintain it, to do so carefully. I think this whole recent episode shows the need to be careful with a production version, and to engage more people in review of development versions. Though I can't produce useful pull requests, I could provide useful input on something that I'm not depending on urgently for publication, but haven't known how to engage with that process. Michael

On 10/03/2016 2:20 PM, Tobie Langel wrote:
On Thu, 10 Mar 2016, at 20:13, Michael Cooper wrote:
On 10/03/2016 2:06 PM, Tobie Langel wrote:
On Thu, 10 Mar 2016, at 19:53, Michael Cooper wrote:
There are still changes impacting publication; this version is adding
"It is expected to become a W3C Note." for a Note-track document that
pubrules rejects; the previous version did not add that, and is accepted
by pubrules. This is the one giving me pain right now but there may be
others. The output of the new respec needs checking against pubrules
before it can be considered fixed. Michael
I'm sure a pull request fixing this would be most welcome. :)
When I'm in publication hell, exacerbated by a recently broken tool, 
trying to learn the code in order to submit a pull request is just not 
possible. I just don't have time to deal with that. 
Remember open-source contributors don't owe you anything and might be as
exacerbated by your expression of entitlement as you might be by their
broken tool. A tool which you are getting for free. :)

--tobie


Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Tobie Langel-4
On Thu, 10 Mar 2016, at 20:38, Michael Cooper wrote:
I'm sorry, I don't mean to sound entitled, but recognize that the wording of my message does sound that way. I know Respec is a volunteer project made available for free. I would like to be able to contribute and share the load, but it uses a technology I just don't have skills in. So I have to depend on the people who do maintain it, to do so carefully. I think this whole recent episode shows the need to be careful with a production version, and to engage more people in review of development versions. Though I can't produce useful pull requests, I could provide useful input on something that I'm not depending on urgently for publication, but haven't known how to engage with that process. Michael
 
Apologies accepted. :)
 
I do think the team is quite aware that this update happened in less than ideal conditions (though again, I can't speak on their behalf).

It might be worth discussing solutions to easily switch to a previous stable version in case the latest release has bugs which block you from publishing (I think Shane kind of informally set something of that sort up recently).
 
--tobie
 
 
 
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec updated

Shane McCarron-6
We are painfully aware that we screwed up a bunch of people.  You have no idea how aware.  

I am going to put new processes in place moving forward.  As to rolling back to stable versions.... yes, we should have done that two weeks ago.  I considered it several times.  I could have pulled a branch off and created a new release with the 1 change needed.  I got wrapped up in the "we are 2 seconds from fixing the core problem" mindset.  For 10 days. I am very upset with myself.  You would think after 35 years in the industry I would know better.

On Thu, Mar 10, 2016 at 1:49 PM, Tobie Langel <[hidden email]> wrote:
On Thu, 10 Mar 2016, at 20:38, Michael Cooper wrote:
I'm sorry, I don't mean to sound entitled, but recognize that the wording of my message does sound that way. I know Respec is a volunteer project made available for free. I would like to be able to contribute and share the load, but it uses a technology I just don't have skills in. So I have to depend on the people who do maintain it, to do so carefully. I think this whole recent episode shows the need to be careful with a production version, and to engage more people in review of development versions. Though I can't produce useful pull requests, I could provide useful input on something that I'm not depending on urgently for publication, but haven't known how to engage with that process. Michael
 
Apologies accepted. :)
 
I do think the team is quite aware that this update happened in less than ideal conditions (though again, I can't speak on their behalf).

It might be worth discussing solutions to easily switch to a previous stable version in case the latest release has bugs which block you from publishing (I think Shane kind of informally set something of that sort up recently).
 
--tobie
 
 
 



--
Shane McCarron
Projects Manager, Spec-Ops