You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@couchdb.apache.org by Alexander Shorin <kx...@gmail.com> on 2014/05/23 18:10:02 UTC

[DISCUSSION] Remove Logs module from Fauxton

Hi devs,

The situation around Logs module looks like the following:

1. Current UX is bad - it requires a lot of work and time to get fixed;
2. After BigCouch merge the /_log resource will be gone (and all the
work that had done for it too);
3. With BigCouch merge lager module comes which is able to streams
logs to custom aggregation systems like LogStash. These systems are
more powerful in question of logs analysis and monitoring than Fauxton
could ever provide;
4. The last release of 1.x series will have LTS status which means
that it'll receive only bug fixes and security updates. Probably, with
one exception in face of Fauxton to make users migration more smooth.
Since Fauxton evolves very fast, it would be awesome to pick his
changes back to LTS Fauxton teams will be happy to reduce amount of
things which only works for some single branch (2.x or 1.x) and the
/_log resource is the only of such.

In overall, there left no strong reasons to continue improve Fauxton
Logs module. May be remove it and close all the related tickets as
Won't Fix? It's a pity to say that, but the reality is hard. What do
you think about?

--
,,,^..^,,,

Re: [DISCUSSION] Remove Logs module from Fauxton

Posted by Robert Kowalski <ro...@kowalski.gd>.
Hi,

I would also just remove the code.


2014-05-24 19:18 GMT+02:00 Simon Metson <si...@cloudant.com>:

> Hey,
> I'd remove the code and let the VCS be the historical record.
> Cheers
> Simon
>
>
> On Saturday, 24 May 2014 at 07:44, Garren Smith wrote:
>
> > Hi All,
> >
> > I’m happy for it to be removed. Should we remove the code or just
> disable from being added to Fauxton on a build? If we keep the code we will
> have something to build off if logs gets added back in after the merge?
> >
> > Cheers
> > Garren
> >
> > On 23 May 2014, at 8:44 PM, Joan Touzet <jo...@lrtw.org> wrote:
> >
> > > Hi all,
> > >
> > > I am also +1 on this. It would be a "nice to have" to restore logging
> > > functionality after the merge, but as the present work is in an
> > > unfinished and unusable state, it would be unfair to users to push this
> > > out.
> > >
> > > -Joan
> > >
> > > ----- Original Message -----
> > > From: "Robert Kowalski" <ro...@kowalski.gd>
> > > To: dev@couchdb.apache.org
> > > Sent: Friday, May 23, 2014 1:13:25 PM
> > > Subject: Re: [DISCUSSION] Remove Logs module from Fauxton
> > >
> > > As the BigCouch merge is right around the corner, I am +1 on closing
> the
> > > tickets for the log component (so people that want to contribute do not
> > > spent time on a component which gets removed in a few weeks). This also
> > > includes not to add features for the log component that do not have a
> > > ticket.
> > >
> > > Also +1 on removing the logs component now, but just if we have to
> support
> > > it with/after 2.0, which would be the case if we do a LTS release. We
> would
> > > have to maintain a special snowflake then.
> > >
> > >
> > >
> > > 2014-05-23 18:10 GMT+02:00 Alexander Shorin <kx...@gmail.com>:
> > >
> > > > Hi devs,
> > > >
> > > > The situation around Logs module looks like the following:
> > > >
> > > > 1. Current UX is bad - it requires a lot of work and time to get
> fixed;
> > > > 2. After BigCouch merge the /_log resource will be gone (and all the
> > > > work that had done for it too);
> > > > 3. With BigCouch merge lager module comes which is able to streams
> > > > logs to custom aggregation systems like LogStash. These systems are
> > > > more powerful in question of logs analysis and monitoring than
> Fauxton
> > > > could ever provide;
> > > > 4. The last release of 1.x series will have LTS status which means
> > > > that it'll receive only bug fixes and security updates. Probably,
> with
> > > > one exception in face of Fauxton to make users migration more smooth.
> > > > Since Fauxton evolves very fast, it would be awesome to pick his
> > > > changes back to LTS Fauxton teams will be happy to reduce amount of
> > > > things which only works for some single branch (2.x or 1.x) and the
> > > > /_log resource is the only of such.
> > > >
> > > > In overall, there left no strong reasons to continue improve Fauxton
> > > > Logs module. May be remove it and close all the related tickets as
> > > > Won't Fix? It's a pity to say that, but the reality is hard. What do
> > > > you think about?
> > > >
> > > > --
> > > > ,,,^..^,,,
> > > >
> > >
> > >
> >
> >
> >
>
>
>

Re: [DISCUSSION] Remove Logs module from Fauxton

Posted by Simon Metson <si...@cloudant.com>.
Hey,  
I'd remove the code and let the VCS be the historical record.  
Cheers
Simon


On Saturday, 24 May 2014 at 07:44, Garren Smith wrote:

> Hi All,
>  
> I’m happy for it to be removed. Should we remove the code or just disable from being added to Fauxton on a build? If we keep the code we will have something to build off if logs gets added back in after the merge?
>  
> Cheers
> Garren
>  
> On 23 May 2014, at 8:44 PM, Joan Touzet <jo...@lrtw.org> wrote:
>  
> > Hi all,
> >  
> > I am also +1 on this. It would be a "nice to have" to restore logging
> > functionality after the merge, but as the present work is in an
> > unfinished and unusable state, it would be unfair to users to push this
> > out.
> >  
> > -Joan
> >  
> > ----- Original Message -----
> > From: "Robert Kowalski" <ro...@kowalski.gd>
> > To: dev@couchdb.apache.org
> > Sent: Friday, May 23, 2014 1:13:25 PM
> > Subject: Re: [DISCUSSION] Remove Logs module from Fauxton
> >  
> > As the BigCouch merge is right around the corner, I am +1 on closing the
> > tickets for the log component (so people that want to contribute do not
> > spent time on a component which gets removed in a few weeks). This also
> > includes not to add features for the log component that do not have a
> > ticket.
> >  
> > Also +1 on removing the logs component now, but just if we have to support
> > it with/after 2.0, which would be the case if we do a LTS release. We would
> > have to maintain a special snowflake then.
> >  
> >  
> >  
> > 2014-05-23 18:10 GMT+02:00 Alexander Shorin <kx...@gmail.com>:
> >  
> > > Hi devs,
> > >  
> > > The situation around Logs module looks like the following:
> > >  
> > > 1. Current UX is bad - it requires a lot of work and time to get fixed;
> > > 2. After BigCouch merge the /_log resource will be gone (and all the
> > > work that had done for it too);
> > > 3. With BigCouch merge lager module comes which is able to streams
> > > logs to custom aggregation systems like LogStash. These systems are
> > > more powerful in question of logs analysis and monitoring than Fauxton
> > > could ever provide;
> > > 4. The last release of 1.x series will have LTS status which means
> > > that it'll receive only bug fixes and security updates. Probably, with
> > > one exception in face of Fauxton to make users migration more smooth.
> > > Since Fauxton evolves very fast, it would be awesome to pick his
> > > changes back to LTS Fauxton teams will be happy to reduce amount of
> > > things which only works for some single branch (2.x or 1.x) and the
> > > /_log resource is the only of such.
> > >  
> > > In overall, there left no strong reasons to continue improve Fauxton
> > > Logs module. May be remove it and close all the related tickets as
> > > Won't Fix? It's a pity to say that, but the reality is hard. What do
> > > you think about?
> > >  
> > > --
> > > ,,,^..^,,,
> > >  
> >  
> >  
>  
>  
>  



Re: [DISCUSSION] Remove Logs module from Fauxton

Posted by Garren Smith <ga...@apache.org>.
Hi All,

I’m happy for it to be removed. Should we remove the code or just disable from being added to Fauxton on a build? If we keep the code we will have something to build off if logs gets added back in after the merge?

Cheers
Garren

On 23 May 2014, at 8:44 PM, Joan Touzet <jo...@lrtw.org> wrote:

> Hi all,
> 
> I am also +1 on this. It would be a "nice to have" to restore logging
> functionality after the merge, but as the present work is in an
> unfinished and unusable state, it would be unfair to users to push this
> out.
> 
> -Joan
> 
> ----- Original Message -----
> From: "Robert Kowalski" <ro...@kowalski.gd>
> To: dev@couchdb.apache.org
> Sent: Friday, May 23, 2014 1:13:25 PM
> Subject: Re: [DISCUSSION] Remove Logs module from Fauxton
> 
> As the BigCouch merge is right around the corner, I am +1 on closing the
> tickets for the log component (so people that want to contribute do not
> spent time on a component which gets removed in a few weeks). This also
> includes not to add features for the log component that do not have a
> ticket.
> 
> Also +1 on removing the logs component now, but just if we have to support
> it with/after 2.0, which would be the case if we do a LTS release. We would
> have to maintain a special snowflake then.
> 
> 
> 
> 2014-05-23 18:10 GMT+02:00 Alexander Shorin <kx...@gmail.com>:
> 
>> Hi devs,
>> 
>> The situation around Logs module looks like the following:
>> 
>> 1. Current UX is bad - it requires a lot of work and time to get fixed;
>> 2. After BigCouch merge the /_log resource will be gone (and all the
>> work that had done for it too);
>> 3. With BigCouch merge lager module comes which is able to streams
>> logs to custom aggregation systems like LogStash. These systems are
>> more powerful in question of logs analysis and monitoring than Fauxton
>> could ever provide;
>> 4. The last release of 1.x series will have LTS status which means
>> that it'll receive only bug fixes and security updates. Probably, with
>> one exception in face of Fauxton to make users migration more smooth.
>> Since Fauxton evolves very fast, it would be awesome to pick his
>> changes back to LTS Fauxton teams will be happy to reduce amount of
>> things which only works for some single branch (2.x or 1.x) and the
>> /_log resource is the only of such.
>> 
>> In overall, there left no strong reasons to continue improve Fauxton
>> Logs module. May be remove it and close all the related tickets as
>> Won't Fix? It's a pity to say that, but the reality is hard. What do
>> you think about?
>> 
>> --
>> ,,,^..^,,,
>> 


Re: [DISCUSSION] Remove Logs module from Fauxton

Posted by Jan Lehnardt <ja...@apache.org>.
On 23 May 2014, at 20:44 , Joan Touzet <jo...@lrtw.org> wrote:

> Hi all,
> 
> I am also +1 on this. It would be a "nice to have" to restore logging
> functionality after the merge, but as the present work is in an
> unfinished and unusable state, it would be unfair to users to push this
> out.

+1 to have this added back in a point release, if we decide we want this.

Best
Jan
--


> 
> -Joan
> 
> ----- Original Message -----
> From: "Robert Kowalski" <ro...@kowalski.gd>
> To: dev@couchdb.apache.org
> Sent: Friday, May 23, 2014 1:13:25 PM
> Subject: Re: [DISCUSSION] Remove Logs module from Fauxton
> 
> As the BigCouch merge is right around the corner, I am +1 on closing the
> tickets for the log component (so people that want to contribute do not
> spent time on a component which gets removed in a few weeks). This also
> includes not to add features for the log component that do not have a
> ticket.
> 
> Also +1 on removing the logs component now, but just if we have to support
> it with/after 2.0, which would be the case if we do a LTS release. We would
> have to maintain a special snowflake then.
> 
> 
> 
> 2014-05-23 18:10 GMT+02:00 Alexander Shorin <kx...@gmail.com>:
> 
>> Hi devs,
>> 
>> The situation around Logs module looks like the following:
>> 
>> 1. Current UX is bad - it requires a lot of work and time to get fixed;
>> 2. After BigCouch merge the /_log resource will be gone (and all the
>> work that had done for it too);
>> 3. With BigCouch merge lager module comes which is able to streams
>> logs to custom aggregation systems like LogStash. These systems are
>> more powerful in question of logs analysis and monitoring than Fauxton
>> could ever provide;
>> 4. The last release of 1.x series will have LTS status which means
>> that it'll receive only bug fixes and security updates. Probably, with
>> one exception in face of Fauxton to make users migration more smooth.
>> Since Fauxton evolves very fast, it would be awesome to pick his
>> changes back to LTS Fauxton teams will be happy to reduce amount of
>> things which only works for some single branch (2.x or 1.x) and the
>> /_log resource is the only of such.
>> 
>> In overall, there left no strong reasons to continue improve Fauxton
>> Logs module. May be remove it and close all the related tickets as
>> Won't Fix? It's a pity to say that, but the reality is hard. What do
>> you think about?
>> 
>> --
>> ,,,^..^,,,
>> 


Re: [DISCUSSION] Remove Logs module from Fauxton

Posted by Joan Touzet <jo...@lrtw.org>.
Hi all,

I am also +1 on this. It would be a "nice to have" to restore logging
functionality after the merge, but as the present work is in an
unfinished and unusable state, it would be unfair to users to push this
out.

-Joan

----- Original Message -----
From: "Robert Kowalski" <ro...@kowalski.gd>
To: dev@couchdb.apache.org
Sent: Friday, May 23, 2014 1:13:25 PM
Subject: Re: [DISCUSSION] Remove Logs module from Fauxton

As the BigCouch merge is right around the corner, I am +1 on closing the
tickets for the log component (so people that want to contribute do not
spent time on a component which gets removed in a few weeks). This also
includes not to add features for the log component that do not have a
ticket.

Also +1 on removing the logs component now, but just if we have to support
it with/after 2.0, which would be the case if we do a LTS release. We would
have to maintain a special snowflake then.



2014-05-23 18:10 GMT+02:00 Alexander Shorin <kx...@gmail.com>:

> Hi devs,
>
> The situation around Logs module looks like the following:
>
> 1. Current UX is bad - it requires a lot of work and time to get fixed;
> 2. After BigCouch merge the /_log resource will be gone (and all the
> work that had done for it too);
> 3. With BigCouch merge lager module comes which is able to streams
> logs to custom aggregation systems like LogStash. These systems are
> more powerful in question of logs analysis and monitoring than Fauxton
> could ever provide;
> 4. The last release of 1.x series will have LTS status which means
> that it'll receive only bug fixes and security updates. Probably, with
> one exception in face of Fauxton to make users migration more smooth.
> Since Fauxton evolves very fast, it would be awesome to pick his
> changes back to LTS Fauxton teams will be happy to reduce amount of
> things which only works for some single branch (2.x or 1.x) and the
> /_log resource is the only of such.
>
> In overall, there left no strong reasons to continue improve Fauxton
> Logs module. May be remove it and close all the related tickets as
> Won't Fix? It's a pity to say that, but the reality is hard. What do
> you think about?
>
> --
> ,,,^..^,,,
>

Re: [DISCUSSION] Remove Logs module from Fauxton

Posted by Robert Kowalski <ro...@kowalski.gd>.
As the BigCouch merge is right around the corner, I am +1 on closing the
tickets for the log component (so people that want to contribute do not
spent time on a component which gets removed in a few weeks). This also
includes not to add features for the log component that do not have a
ticket.

Also +1 on removing the logs component now, but just if we have to support
it with/after 2.0, which would be the case if we do a LTS release. We would
have to maintain a special snowflake then.



2014-05-23 18:10 GMT+02:00 Alexander Shorin <kx...@gmail.com>:

> Hi devs,
>
> The situation around Logs module looks like the following:
>
> 1. Current UX is bad - it requires a lot of work and time to get fixed;
> 2. After BigCouch merge the /_log resource will be gone (and all the
> work that had done for it too);
> 3. With BigCouch merge lager module comes which is able to streams
> logs to custom aggregation systems like LogStash. These systems are
> more powerful in question of logs analysis and monitoring than Fauxton
> could ever provide;
> 4. The last release of 1.x series will have LTS status which means
> that it'll receive only bug fixes and security updates. Probably, with
> one exception in face of Fauxton to make users migration more smooth.
> Since Fauxton evolves very fast, it would be awesome to pick his
> changes back to LTS Fauxton teams will be happy to reduce amount of
> things which only works for some single branch (2.x or 1.x) and the
> /_log resource is the only of such.
>
> In overall, there left no strong reasons to continue improve Fauxton
> Logs module. May be remove it and close all the related tickets as
> Won't Fix? It's a pity to say that, but the reality is hard. What do
> you think about?
>
> --
> ,,,^..^,,,
>