Re: [Bloat] bufferbloat.net is sorely missed.

Re: [Bloat] bufferbloat.net is sorely missed.

[Rich Brown]

Toke,

Thanks for taking on the revival of the bufferbloat site. 

As a stopgap, I wonder if it might be possible to put up a placeholder page that links to Wayback Machine pages. Something like... http://richb-hanover.github.io/bufferbloat-site/ (I have thrown $50 toward the archive.org folks...) It would be easy to redirect the www.bufferbloat.net URL to this site.

As a secondary step, we might consider using github.io pages with Jekyll. I just converted another static site to Jekyll with astonishing ease. It's now editable with Markdown on github. 

If you have the text of those pages in some sensible directory structure, I bet it would be trivial to make a Jekyll site.

Let me know if this helps.

Rich

> On Jun 10, 2016, at 7:57 AM, bloat-request@lists.bufferbloat.net wrote:
> 
> The Wayback Machine is your friend:
> 
> https://web.archive.org/web/20160328150326/http://www.bufferbloat.net/projects/codel/wiki/Cake

Re: [Bloat] bufferbloat.net is sorely missed.

[Dave Taht]

On Fri, Jun 10, 2016 at 8:27 AM, Rich Brown <richb.hanover@gmail.com> wrote:
> Toke,
>
> Thanks for taking on the revival of the bufferbloat site.

While I am delighted by progress here, given a choice between landing
more fixes to wifi and fixing the site, I'd rather we keep building on
the momentum for fixing wifi. With a core bit landing in net-next
soon, I hope, and airtime fairness well on it's way, the possibility
exists to get kernel 4.8 or 4.9 truly being a vast improvement on wifi
for multiple devices building on the finicy bits of that
infrastructure.

that said toke seems to be achieving $DIETY-like results across many areas.

> As a stopgap, I wonder if it might be possible to put up a placeholder page that links to Wayback Machine pages. Something like... http://richb-hanover.github.io/bufferbloat-site/ (I have thrown $50 toward the archive.org folks...) It would be easy to redirect the www.bufferbloat.net URL to this site.

archive.org has an automagic 404 converter, if that helps.

https://blog.archive.org/2013/10/24/web-archive-404-handler-for-webmasters/

>
> As a secondary step, we might consider using github.io pages with Jekyll. I just converted another static site to Jekyll with astonishing ease. It's now editable with Markdown on github.

Hugo vastly outperforms jekyll, has a jekyll converter, and I'd prefer
to stick with hugo.

In the long run I would like to also resurrect your work on converting
lartc.org.

>
> If you have the text of those pages in some sensible directory structure, I bet it would be trivial to make a Jekyll site.

Yes, I'd like the directory structure to more or less match what the urls were.


>
> Let me know if this helps.
>
> Rich
>
>> On Jun 10, 2016, at 7:57 AM, bloat-request@lists.bufferbloat.net wrote:
>>
>> The Wayback Machine is your friend:
>>
>> https://web.archive.org/web/20160328150326/http://www.bufferbloat.net/projects/codel/wiki/Cake
>
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Bloat] bufferbloat.net is sorely missed.

[Rich Brown]

Hi Dave,

I'm worried that the bufferbloat project appears MIA right now.

It won't take much time to get that github page (http://richb-hanover.github.io/bufferbloat-site) running as www.bufferbloat.net. To do this:

- You would have to set up A records for bufferbloat.net to point to 192.30.252.153 and 192.30.252.154
- Wildcard address (for www.bufferbloat.net) should CNAME to the 'bufferbloat.net' A record
- I would add a file (named 'CNAME') to the gh-pages branch of the github repo at https://github.com/richb-hanover/bufferbloat-site

One DNS TTL later, we should have a bufferbloat.net web presence back on the air.

Longer term, I don't have a preference about the web publishing system: I just want us to appear alive. Thanks.

Rich

Re: [Bloat] bufferbloat.net is sorely missed.

[Kevin Darbyshire-Bryant]

On 10/06/16 17:19, Rich Brown wrote:
> Hi Dave,
>
> I'm worried that the bufferbloat project appears MIA right now.
Absolutely.  And the sqm-scripts also mention going to bufferbloat.net, 
so hitting a 404 is underwhelming, especially as LEDE users have no 
'CAKE' man pages to which to refer.
>
> It won't take much time to get that github page (http://richb-hanover.github.io/bufferbloat-site) running as www.bufferbloat.net. To do this:
>
> - You would have to set up A records for bufferbloat.net to point to 192.30.252.153 and 192.30.252.154
> - Wildcard address (for www.bufferbloat.net) should CNAME to the 'bufferbloat.net' A record
> - I would add a file (named 'CNAME') to the gh-pages branch of the github repo at https://github.com/richb-hanover/bufferbloat-site
>
> One DNS TTL later, we should have a bufferbloat.net web presence back on the air.
>
> Longer term, I don't have a preference about the web publishing system: I just want us to appear alive. Thanks.
Agree 100%
>
> Rich
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] bufferbloat.net is sorely missed.

[Outback Dingo]

[-- Attachment #1: Type: text/plain, Size: 1169 bytes --]

On Fri, Jun 10, 2016 at 12:19 PM, Rich Brown <richb.hanover@gmail.com>
wrote:

> Hi Dave,
>
> I'm worried that the bufferbloat project appears MIA right now.
>
> It won't take much time to get that github page (
> http://richb-hanover.github.io/bufferbloat-site) running as
> www.bufferbloat.net. To do this:
>
> - You would have to set up A records for bufferbloat.net to point to
> 192.30.252.153 and 192.30.252.154
> - Wildcard address (for www.bufferbloat.net) should CNAME to the '
> bufferbloat.net' A record
> - I would add a file (named 'CNAME') to the gh-pages branch of the github
> repo at https://github.com/richb-hanover/bufferbloat-site
>
> One DNS TTL later, we should have a bufferbloat.net web presence back on
> the air.
>
> Longer term, I don't have a preference about the web publishing system: I
> just want us to appear alive. Thanks.
>

Is it that bufferbloat.net requires a permanent home ?? I can provide a
decent sized vm and dns at 0 cost for hosting the site
on my servers......



>
> Rich
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>

[-- Attachment #2: Type: text/html, Size: 2553 bytes --]

Re: [Bloat] bufferbloat.net is sorely missed.

[Dave Taht]

On Fri, Jun 10, 2016 at 10:13 AM, Outback Dingo <outbackdingo@gmail.com> wrote:
>
>
> On Fri, Jun 10, 2016 at 12:19 PM, Rich Brown <richb.hanover@gmail.com>
> wrote:
>>
>> Hi Dave,
>>
>> I'm worried that the bufferbloat project appears MIA right now.
>>
>> It won't take much time to get that github page
>> (http://richb-hanover.github.io/bufferbloat-site) running as
>> www.bufferbloat.net. To do this:
>>
>> - You would have to set up A records for bufferbloat.net to point to
>> 192.30.252.153 and 192.30.252.154
>> - Wildcard address (for www.bufferbloat.net) should CNAME to the
>> 'bufferbloat.net' A record
>> - I would add a file (named 'CNAME') to the gh-pages branch of the github
>> repo at https://github.com/richb-hanover/bufferbloat-site
>>
>> One DNS TTL later, we should have a bufferbloat.net web presence back on
>> the air.

OK, tomorrow (or tonight) I will look into the redirect rich suggests.
I note several other folk do have access to the dns server and if they
beat me to it I won't mind at all....

Toke has a preliminary conversion done but it lacks fixing the paths
and bug issues as well (archive.org is also missing the issues db).

https://kau.toke.dk/bufferbloat-archive/wiki/

>> Longer term, I don't have a preference about the web publishing system: I
>> just want us to appear alive. Thanks.

I am only alive for small values of alive. I am glad "us" is still
alive and kicking.

>
> Is it that bufferbloat.net requires a permanent home ?? I can provide a
> decent sized vm and dns at 0 cost for hosting the site
> on my servers......

Well, if I had any advanced plan, it was to convert to a static, git
managed, hugo repo, and start using a CDN of some sort directly. I
have plenty of linodes lying around, but if we could get worldwide
distribution for a (cdns seem to charge for data transfer rather than
storage), without even having a central site, that would be nice.

Secondarily I'd really like to transfer all the fixed assets I
maintain and pay for to some independent set of individuals or an
institution (nlnet's upcoming "commons conservancy" is the strongest
possibility - https://nlnet.nl/project/commonsconservancy/ for some
details) - or adopt something like lede's "3 man rule" for access to
all assets.

>
>
>>
>>
>> Rich
>> _______________________________________________
>> Bloat mailing list
>> Bloat@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/bloat
>
>
>
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Bloat] bufferbloat.net is sorely missed.

[Steinar H. Gunderson]

On Fri, Jun 10, 2016 at 11:05:29AM -0700, Dave Taht wrote:
> Secondarily I'd really like to transfer all the fixed assets I
> maintain and pay for to some independent set of individuals or an
> institution (nlnet's upcoming "commons conservancy" is the strongest
> possibility - https://nlnet.nl/project/commonsconservancy/ for some
> details) - or adopt something like lede's "3 man rule" for access to
> all assets.

Try talking to SPI (Software in the Public Interest)?

/* Steinar */
-- 
Homepage: https://www.sesse.net/

Re: [Bloat] bufferbloat.net is sorely missed.

[Rich Brown]

OK. Let me know when the DNS has been updated, and I will put in the CNAME file.  Thanks!

Rich

> On Jun 10, 2016, at 2:05 PM, Dave Taht <dave.taht@gmail.com> wrote:
> 
>>> It won't take much time to get that github page
>>> (http://richb-hanover.github.io/bufferbloat-site) running as
>>> www.bufferbloat.net. To do this:
>>> 
>>> - You would have to set up A records for bufferbloat.net to point to
>>> 192.30.252.153 and 192.30.252.154
>>> - Wildcard address (for www.bufferbloat.net) should CNAME to the
>>> 'bufferbloat.net' A record
>>> - I would add a file (named 'CNAME') to the gh-pages branch of the github
>>> repo at https://github.com/richb-hanover/bufferbloat-site
>>> 
>>> One DNS TTL later, we should have a bufferbloat.net web presence back on
>>> the air.
> 
> OK, tomorrow (or tonight) I will look into the redirect rich suggests.
> I note several other folk do have access to the dns server and if they
> beat me to it I won't mind at all....

Re: [Bloat] bufferbloat.net is sorely missed.

[Dave Taht]

On Fri, Jun 10, 2016 at 11:05 AM, Dave Taht <dave.taht@gmail.com> wrote:
> On Fri, Jun 10, 2016 at 10:13 AM, Outback Dingo <outbackdingo@gmail.com> wrote:
>>
>>
>> On Fri, Jun 10, 2016 at 12:19 PM, Rich Brown <richb.hanover@gmail.com>
>> wrote:
>>>
>>> Hi Dave,
>>>
>>> I'm worried that the bufferbloat project appears MIA right now.
>>>
>>> It won't take much time to get that github page
>>> (http://richb-hanover.github.io/bufferbloat-site) running as
>>> www.bufferbloat.net. To do this:
>>>
>>> - You would have to set up A records for bufferbloat.net to point to
>>> 192.30.252.153 and 192.30.252.154
>>> - Wildcard address (for www.bufferbloat.net) should CNAME to the
>>> 'bufferbloat.net' A record
>>> - I would add a file (named 'CNAME') to the gh-pages branch of the github
>>> repo at https://github.com/richb-hanover/bufferbloat-site
>>>
>>> One DNS TTL later, we should have a bufferbloat.net web presence back on
>>> the air.
>
> OK, tomorrow (or tonight) I will look into the redirect rich suggests.
> I note several other folk do have access to the dns server and if they
> beat me to it I won't mind at all....
>
> Toke has a preliminary conversion done but it lacks fixing the paths
> and bug issues as well (archive.org is also missing the issues db).
>
> https://kau.toke.dk/bufferbloat-archive/wiki/
>
>>> Longer term, I don't have a preference about the web publishing system: I
>>> just want us to appear alive. Thanks.
>
> I am only alive for small values of alive. I am glad "us" is still
> alive and kicking.
>
>>
>> Is it that bufferbloat.net requires a permanent home ?? I can provide a
>> decent sized vm and dns at 0 cost for hosting the site
>> on my servers......
>
> Well, if I had any advanced plan, it was to convert to a static, git
> managed, hugo repo, and start using a CDN of some sort directly. I
> have plenty of linodes lying around, but if we could get worldwide
> distribution for a (cdns seem to charge for data transfer rather than
> storage), without even having a central site, that would be nice.

For example, cloudfront has "50 GB Data Transfer Out, 2,000,000 HTTP
and HTTPS Requests", for free for a year.

https://aws.amazon.com/free/

I sincerely doubt www.bufferbloat.net would do more than that.

I haven't looked into other CDNs and haven't the foggiest idea how to
set one up in the first place.

Presently I have going out $150/mo in linodes (mostly flent servers, 2
being used for web pages), $250/mo in the google cloud for openwrt's
build bot, probably $500/mo for lede's build bot...

and $190/mo coming in. I do hope to find more funding for lede's
stuff... (or get out of it after they've been better bootstrapped)

> Secondarily I'd really like to transfer all the fixed assets I
> maintain and pay for to some independent set of individuals or an
> institution (nlnet's upcoming "commons conservancy" is the strongest
> possibility - https://nlnet.nl/project/commonsconservancy/ for some
> details) - or adopt something like lede's "3 man rule" for access to
> all assets.
>
>>
>>
>>>
>>>
>>> Rich
>>> _______________________________________________
>>> Bloat mailing list
>>> Bloat@lists.bufferbloat.net
>>> https://lists.bufferbloat.net/listinfo/bloat
>>
>>
>>
>> _______________________________________________
>> Bloat mailing list
>> Bloat@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/bloat
>>
>
>
>
> --
> Dave Täht
> Let's go make home routers and wifi faster! With better software!
> http://blog.cerowrt.org



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Bloat] bufferbloat.net is sorely missed.

[Toke Høiland-Jørgensen]

Rich Brown <richb.hanover@gmail.com> writes:

> Thanks for taking on the revival of the bufferbloat site. 

Okay, updated the conversion to preserve the directory structure.
Current version is here: https://kau.toke.dk/bufferbloat-archive. The
overview pages and index are somewhat lacking at the moment, but the
wiki pages are there at the same relative URLs as on the old site.

I can update the DNS to point at that server and have old links start
working again; if no one objects, I'll do that later today.

There's a github repo with the contents linked from the above link; pull
requests welcome if anyone wants to contribute to improving things :)

-Toke

Re: [Bloat] bufferbloat.net is sorely missed.

[Dave Taht]

On Sat, Jun 11, 2016 at 5:07 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Rich Brown <richb.hanover@gmail.com> writes:
>
>> Thanks for taking on the revival of the bufferbloat site.
>
> Okay, updated the conversion to preserve the directory structure.
> Current version is here: https://kau.toke.dk/bufferbloat-archive. The

I get a 404 on that right now across the board...

https://kau.toke.dk/bufferbloat-archive/projects/codel/wiki/Benchmarking_Codel_and_FQ_Codel/


> overview pages and index are somewhat lacking at the moment, but the
> wiki pages are there at the same relative URLs as on the old site.

Having the original data available to those willing to work on fixing
it would be useful, I was reluctant to share the sql dump widely due
to passwords being in it. Can more of the sql dump land somewhere? Do
you have other conversion scripts lying around?

(I used pandoc to convert from textile to markdown originally plus a
few sed scripts)

> I can update the DNS to point at that server and have old links start
> working again; if no one objects, I'll do that later today.

Well, to save on cost and improve bandwidth I'd rather it moved to the
web server lists.bufferbloat.net is on. (and ultimately a cdn). I was
reluctant to move it there because the redmine integration was
failing, but for a static web site, no problem.

...

For now, feel free to slam yourself. :). I'd done a few things to
improve caching behavior on the blog.cerowrt.org apache
implementation, also.

> There's a github repo with the contents linked from the above link; pull
> requests welcome if anyone wants to contribute to improving things :)

I am personally slammed and delighted others are stepping up to take
care of this...

> -Toke
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Bloat] bufferbloat.net is sorely missed.

[Toke Høiland-Jørgensen]

Dave Taht <dave.taht@gmail.com> writes:

>> Okay, updated the conversion to preserve the directory structure.
>> Current version is here: https://kau.toke.dk/bufferbloat-archive. The
>
> I get a 404 on that right now across the board...
>
> https://kau.toke.dk/bufferbloat-archive/projects/codel/wiki/Benchmarking_Codel_and_FQ_Codel/

Was moving things around and hadn't set up a redirect yet. Should work
now, and the server also answers to www.bufferbloat.net. I did an
nsupdate, but the linode nameservers don't seem to have picked it up
yet. Not sure how to poke them to do so.

>> overview pages and index are somewhat lacking at the moment, but the
>> wiki pages are there at the same relative URLs as on the old site.
>
> Having the original data available to those willing to work on fixing
> it would be useful, I was reluctant to share the sql dump widely due
> to passwords being in it. Can more of the sql dump land somewhere? Do
> you have other conversion scripts lying around?

This is all in the github repo in the export/ dir - both scripts and csv
files. Did another export from the database, merging several tables with
the information needed for the wiki. There's also an export of the
attachments which I haven't done anything with yet.

> (I used pandoc to convert from textile to markdown originally plus a
> few sed scripts)

Yes, mine are python scripts that also run pandoc.

>> I can update the DNS to point at that server and have old links start
>> working again; if no one objects, I'll do that later today.
>
> Well, to save on cost and improve bandwidth I'd rather it moved to the
> web server lists.bufferbloat.net is on. (and ultimately a cdn). I was
> reluctant to move it there because the redmine integration was
> failing, but for a static web site, no problem.

Have no objections to moving it once we hit something a bit more stable;
but for now, having it on my own server allows me to iterate quicker.
It's at my uni on a gbit link, so should be fine (famous last words).

Having several mirrors and sticking all the addresses in DNS would also
be a way to do things, if we can make sure they all update from git
regularly. Figure these things can be sorted out once we get the
conversion itself done and have it go into maintenance mode (where we
actually take pull requests for the *content*). :)

-Toke

Re: [Bloat] bufferbloat.net is sorely missed.

[Toke Høiland-Jørgensen]

Toke Høiland-Jørgensen <toke@toke.dk> writes:

> I did an nsupdate, but the linode nameservers don't seem to have
> picked it up yet. Not sure how to poke them to do so.

Ah, they seem to be switching over now...

-Toke

Re: [Bloat] bufferbloat.net is sorely missed.

[Dave Taht]

for those of you that are not used to the wonderfulness of using hugo
to develop web content, it is amazing.

Install hugo from:  https://gohugo.io/

do a git clone of toke's repo.
cd into that dir.

hugo -D server

It currently on my hardware generates the entire site statically in
490ms (note that as best I recall we had 360 wiki pages not 188, btw),
and you can go and edit anything until you are satisified with it with
one editor and one web window open.

There is a great deal of curation that can be done (after it
stablizes), and also it can be easily checked for errors with tools
like html linkcheckers, etc.

It also loads tons faster than the redmine version.

d@dancer:~/git/bufferbloat-net$ hugo -D server
0 draft content
0 future content
188 pages created
0 non-page files copied
0 paginator pages created
0 tags created
0 categories created
in 490 ms
Watching for changes in /home/d/git/bufferbloat-net/{content,layouts,static}
Serving pages from memory
Web Server is available at http://localhost:1313/projects/


On Sat, Jun 11, 2016 at 11:32 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Toke Høiland-Jørgensen <toke@toke.dk> writes:
>
>> I did an nsupdate, but the linode nameservers don't seem to have
>> picked it up yet. Not sure how to poke them to do so.
>
> Ah, they seem to be switching over now...
>
> -Toke



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Bloat] bufferbloat.net is sorely missed/Now it's back!

[Rich Brown]

Hi Toke,

> On Jun 11, 2016, at 2:29 PM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> 
>>> 
>>> Okay, updated the conversion to preserve the directory structure.
>>> Current version is here: https://kau.toke.dk/bufferbloat-archive. The
>> 
>> I get a 404 on that right now across the board...
>> 
>> https://kau.toke.dk/bufferbloat-archive/projects/codel/wiki/Benchmarking_Codel_and_FQ_Codel/
> 
> Was moving things around and hadn't set up a redirect yet. Should work
> now, and the server also answers to www.bufferbloat.net. I did an
> nsupdate, but the linode nameservers don't seem to have picked it up
> yet. Not sure how to poke them to do so.

The new site just went live here in the shrubs of New Hampshire, USA. (It wasn't working ~10 minutes ago.)

WOW! This is a wonderful piece of work! It manages to regain most of the structure of the separate parts of the Redmine system. 

I just forked the repo and will start looking for places to make it better. Thanks!

Rich

PS I will leave my placeholder github.io site up in case anyone wants to go back via the Wayback Machine. http://richb-hanover.github.io/bufferbloat-site/

Re: [Bloat] bufferbloat.net is sorely missed/Now it's back!

[Toke Høiland-Jørgensen]

Rich Brown <richb.hanover@gmail.com> writes:

> The new site just went live here in the shrubs of New Hampshire, USA.
> (It wasn't working ~10 minutes ago.)

Awesome! The letsencrypt servers seem to have picked up the change too,
so there's now a cert in place and non-https redirects to serve via
https.

> WOW! This is a wonderful piece of work! It manages to regain most of
> the structure of the separate parts of the Redmine system.

Thanks! Yeah, tried to keep the structure from the Redmine install. The
only thing that may not be too obvious from the git repo is that the
static site uses the /project prefix as its root. This is to take
advantage of the built-in 'section' functionality of Hugo. The Hugo
output is simply put in a subfolder on the server, and I configured it
to redirect / to /projects.

> I just forked the repo and will start looking for places to make it
> better. Thanks!

Awesome! Will update the README with a few pointers on how it's
structured (unless you beat me to it ;)).

> PS I will leave my placeholder github.io site up in case anyone wants
> to go back via the Wayback Machine.
> http://richb-hanover.github.io/bufferbloat-site/

Cool. Also added the archive.org 404 helper to the 404 page for the new
site :)


-Toke

Re: [Bloat] bufferbloat.net is sorely missed/Now it's back!

[Dave Taht]

And I see I'm a release behind. I did find merging up from prior
releases to be somewhat painful (as the default templates tended to
change), but I upgraded to .16 with no issues so far.

https://github.com/spf13/hugo/releases


On Sat, Jun 11, 2016 at 1:05 PM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Rich Brown <richb.hanover@gmail.com> writes:
>
>> The new site just went live here in the shrubs of New Hampshire, USA.
>> (It wasn't working ~10 minutes ago.)
>
> Awesome! The letsencrypt servers seem to have picked up the change too,
> so there's now a cert in place and non-https redirects to serve via
> https.
>
>> WOW! This is a wonderful piece of work! It manages to regain most of
>> the structure of the separate parts of the Redmine system.
>
> Thanks! Yeah, tried to keep the structure from the Redmine install. The
> only thing that may not be too obvious from the git repo is that the
> static site uses the /project prefix as its root. This is to take
> advantage of the built-in 'section' functionality of Hugo. The Hugo
> output is simply put in a subfolder on the server, and I configured it
> to redirect / to /projects.
>
>> I just forked the repo and will start looking for places to make it
>> better. Thanks!
>
> Awesome! Will update the README with a few pointers on how it's
> structured (unless you beat me to it ;)).
>
>> PS I will leave my placeholder github.io site up in case anyone wants
>> to go back via the Wayback Machine.
>> http://richb-hanover.github.io/bufferbloat-site/
>
> Cool. Also added the archive.org 404 helper to the 404 page for the new
> site :)
>
>
> -Toke



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

[Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Rich Brown]

To all the Bufferbloaters out there...

Thanks again to Toke for all this good work. It's terrific to see the pages resurrected, and to have the www.bufferbloat.net site living again.

I've been thinking of the next steps - to make all this work really useful. I see there are three important issues:

Organizing: Thinking about the best way to display the current information:
	- Should Cake be listed as a top-level project?
	- Is there a way to display newest entries? I'd like to see a sidebar element showing the 5 newest posts
	- Can an RSS feed be generated automatically?

Curating: The current site makes it seem as if all pages are equally important. I feel the urge to do the following:
	- Categorizing the "List of Wiki Pages" for a project (e.g., http://bufferbloat.net/projects/cerowrt/ ) 
		so there's a sense of their importance
	- Discard old/outdated/useless articles
	
Workflow:
	- Rules for making posts: Who can make them? How do they get published?
	- Auto-publish - if it's not already happening, could we re-render and publish after a commit?
	- I see the 404 handler in place, but it doesn't seem to show a link to archive.org
	- Is there a way to do page redirection? It looks as if Toke has done a great job of replicating
		URLs in the new site, but if we see frequent broken links, is there a way to redirect a page to its equivalent on the new site?
	- Should we add a Google search box on 404 page? On every page?
	- Would it make sense to review error logs from time to time?

I'd love to have your thoughts. Thanks.

Rich

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Toke Høiland-Jørgensen]

Rich Brown <richb.hanover@gmail.com> writes:

> Thanks again to Toke for all this good work. It's terrific to see the
> pages resurrected, and to have the www.bufferbloat.net site living
> again.

You're welcome! Just glad it's being appreciated ;)

> I've been thinking of the next steps - to make all this work really useful. I see there are three important issues:
>
> Organizing: Thinking about the best way to display the current information:
> 	- Should Cake be listed as a top-level project?

No opinion.

> 	- Is there a way to display newest entries? I'd like to see a
> 	sidebar element showing the 5 newest posts

Yes, certainly. Was thinking of putting a list of newest news items on
the front page. However, for the wiki pages it is not necessarily
obvious that having a notion of 'newest' is useful.

> 	- Can an RSS feed be generated automatically?

Yes, but see above.

> Curating: The current site makes it seem as if all pages are equally important. I feel the urge to do the following:
> 	- Categorizing the "List of Wiki Pages" for a project (e.g., http://bufferbloat.net/projects/cerowrt/ ) 
> 		so there's a sense of their importance

A way to do this would be to update the 'index' wiki pages to be more
useful (e.g. https://www.bufferbloat.net/projects/cerowrt/wiki/) and
keep the overview pages as a 'complete' list. However, it's certainly
also possible to mark some of the pages as 'important' and have those
show up on top of the list.


> 	- Discard old/outdated/useless articles

Yes, this is probably needed. And also fixing broken links (see the
issue I opened for this on github).

> Workflow:
> 	- Rules for making posts: Who can make them? How do they get
> 	published?

Well, it's a git repo. I figure anyone can do pull requests.

> 	- Auto-publish - if it's not already happening, could we
> 	re-render and publish after a commit?

Yes, that is certainly possible. For now I figure I'll do it manually
(it's just re-running a script), but if we get enough activity that that
becomes a bottleneck, I can certainly set up something automated.

> 	- I see the 404 handler in place, but it doesn't seem to show a
> 	link to archive.org

It includes a javascript from archive.org which *should* show a link if
(and only if) an archived version exists. Haven't verified that it
actually works.

> 	- Is there a way to do page redirection? It looks as if Toke has done a great job of replicating
> 		URLs in the new site, but if we see frequent broken
> 		links, is there a way to redirect a page to its
> 		equivalent on the new site?

Yes. This is already used for news items and issues.

> 	- Should we add a Google search box on 404 page? On every page?

Not sure what it takes to do this properly, and if it's worth the
hassle.

> 	- Would it make sense to review error logs from time to time?

Possible. Split out the logs to a separate file; can provide it if
someone wants to go digging. Or I can just parse out a list of 404
errors from time to time.

-Toke

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[moeller0]

So how do I contribute changes to existing pages? The cake page has some inaccuracies regarding VDSL2 encapsulation that I would like to fix…

“Apparently PTM does have a small additional overhead on the order of 1⁄128, due to HDLC framing which attaches special meaning to 0x7D and 0x7E bytes; I might need to add approximate handling for that, kernel-side.”

The point is only VDLS1 used HDLC on the relevant data bearers, VDSL2 uses 64/65 encoding that has not reserved octets and hence does not have additional overhead nor magic bytes… (full disclosure VDSL2 still allows HDLC bearers, but just not the one bearer the user actually sends data over, so the HDLC thing is a distraction that we can avoid.)

Best Regards
	Sebastian



> On Jun 12, 2016, at 20:02 , Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> 
> Rich Brown <richb.hanover@gmail.com> writes:
> 
>> Thanks again to Toke for all this good work. It's terrific to see the
>> pages resurrected, and to have the www.bufferbloat.net site living
>> again.
> 
> You're welcome! Just glad it's being appreciated ;)
> 
>> I've been thinking of the next steps - to make all this work really useful. I see there are three important issues:
>> 
>> Organizing: Thinking about the best way to display the current information:
>> 	- Should Cake be listed as a top-level project?
> 
> No opinion.
> 
>> 	- Is there a way to display newest entries? I'd like to see a
>> 	sidebar element showing the 5 newest posts
> 
> Yes, certainly. Was thinking of putting a list of newest news items on
> the front page. However, for the wiki pages it is not necessarily
> obvious that having a notion of 'newest' is useful.
> 
>> 	- Can an RSS feed be generated automatically?
> 
> Yes, but see above.
> 
>> Curating: The current site makes it seem as if all pages are equally important. I feel the urge to do the following:
>> 	- Categorizing the "List of Wiki Pages" for a project (e.g., http://bufferbloat.net/projects/cerowrt/ ) 
>> 		so there's a sense of their importance
> 
> A way to do this would be to update the 'index' wiki pages to be more
> useful (e.g. https://www.bufferbloat.net/projects/cerowrt/wiki/) and
> keep the overview pages as a 'complete' list. However, it's certainly
> also possible to mark some of the pages as 'important' and have those
> show up on top of the list.
> 
> 
>> 	- Discard old/outdated/useless articles
> 
> Yes, this is probably needed. And also fixing broken links (see the
> issue I opened for this on github).
> 
>> Workflow:
>> 	- Rules for making posts: Who can make them? How do they get
>> 	published?
> 
> Well, it's a git repo. I figure anyone can do pull requests.
> 
>> 	- Auto-publish - if it's not already happening, could we
>> 	re-render and publish after a commit?
> 
> Yes, that is certainly possible. For now I figure I'll do it manually
> (it's just re-running a script), but if we get enough activity that that
> becomes a bottleneck, I can certainly set up something automated.
> 
>> 	- I see the 404 handler in place, but it doesn't seem to show a
>> 	link to archive.org
> 
> It includes a javascript from archive.org which *should* show a link if
> (and only if) an archived version exists. Haven't verified that it
> actually works.
> 
>> 	- Is there a way to do page redirection? It looks as if Toke has done a great job of replicating
>> 		URLs in the new site, but if we see frequent broken
>> 		links, is there a way to redirect a page to its
>> 		equivalent on the new site?
> 
> Yes. This is already used for news items and issues.
> 
>> 	- Should we add a Google search box on 404 page? On every page?
> 
> Not sure what it takes to do this properly, and if it's worth the
> hassle.
> 
>> 	- Would it make sense to review error logs from time to time?
> 
> Possible. Split out the logs to a separate file; can provide it if
> someone wants to go digging. Or I can just parse out a list of 404
> errors from time to time.
> 
> -Toke
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Toke Høiland-Jørgensen]

moeller0 <moeller0@gmx.de> writes:

> So how do I contribute changes to existing pages? The cake page has
> some inaccuracies regarding VDSL2 encapsulation that I would like to
> fix…

Clone the repo at https://github.com/tohojo/bufferbloat-net/, make
changes to the right markdown file and submit a pull request with the
change :)

-Toke

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[moeller0]

Hi Toke,



> On Jun 12, 2016, at 20:16 , Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> 
> moeller0 <moeller0@gmx.de> writes:
> 
>> So how do I contribute changes to existing pages? The cake page has
>> some inaccuracies regarding VDSL2 encapsulation that I would like to
>> fix…
> 
> Clone the repo at https://github.com/tohojo/bufferbloat-net/, make
> changes to the right markdown file and submit a pull request with the
> change :)

	Thanks, I just saw this information is also on the bottom of the entry page, sorry for not looking more closely…

Best Regards
	Sebastian

> 
> -Toke

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Toke Høiland-Jørgensen]

moeller0 <moeller0@gmx.de> writes:

>> Clone the repo at https://github.com/tohojo/bufferbloat-net/, make
>> changes to the right markdown file and submit a pull request with the
>> change :)
>
> 	Thanks, I just saw this information is also on the bottom of the
> 	entry page, sorry for not looking more closely…

If you missed it, it's probably not visible enough... ;)

-Toke

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Dave Taht]

On Sun, Jun 12, 2016 at 11:16 AM, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> moeller0 <moeller0@gmx.de> writes:
>
>> So how do I contribute changes to existing pages? The cake page has
>> some inaccuracies regarding VDSL2 encapsulation that I would like to
>> fix…
>
> Clone the repo at https://github.com/tohojo/bufferbloat-net/, make
> changes to the right markdown file and submit a pull request with the
> change :)

Arguably it will be faster as expertise is gained to allow more (most
or all) of the core authors direct access to the primary github repo,
and autopublication is possible with a post-push hook.

It does help to run hugo locally to preview your fixes as you go along.

One of the big "fixes" inherent in this design is that hugo supports a
"draft" concept, where stuff can be iterated on inside of git by one
or more parties before it makes it to the public web. I have a problem
in that I'm always working on 6 things at once and have had a tendency
to realize "oh, that's a topic that needs more work", and yet never
get around to fleshing out that topic, in part, because I forget it's
essentially a broken link.

I envy those that can sit down on a conventional wiki, start, and two
hours later write "the end" and move on. That process for me takes
days, weeks, (or years!). I really liked adopting the bloggy format
for blog.cerowrt.org, as my natural metíer is the rant. I must have
30+ more articles in draft form in git at the moment, there.

> -Toke
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Toke Høiland-Jørgensen]

Dave Taht <dave.taht@gmail.com> writes:

>> Clone the repo at https://github.com/tohojo/bufferbloat-net/, make
>> changes to the right markdown file and submit a pull request with the
>> change :)
>
> Arguably it will be faster as expertise is gained to allow more (most
> or all) of the core authors direct access to the primary github repo,
> and autopublication is possible with a post-push hook.

No objection to that :)

> One of the big "fixes" inherent in this design is that hugo supports a
> "draft" concept, where stuff can be iterated on inside of git by one
> or more parties before it makes it to the public web. I have a problem
> in that I'm always working on 6 things at once and have had a tendency
> to realize "oh, that's a topic that needs more work", and yet never
> get around to fleshing out that topic, in part, because I forget it's
> essentially a broken link.

Well, isn't the idea of a Wiki that you just publish the draft, and it's
always a work in progress? ;)

-Toke

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Rich Brown]

Good comments all... Let me preface these responses by saying that I'm willing to help out, but don't want to get in the way of the rapid progress being made. 

Toke: Let us know when you think you're done with major reorganizations...

I also want to see if there's consensus about when and how to make updates, so that we're not stepping on toes. TL;DR questions:

- Do we only want to use pull requests, or should core authors have edit access to github?
- Or do we use the 'draft' facility? (How?)


>> 	- Is there a way to display newest entries? I'd like to see a
>> 	sidebar element showing the 5 newest posts
> 
> Yes, certainly. Was thinking of putting a list of newest news items on
> the front page. However, for the wiki pages it is not necessarily
> obvious that having a notion of 'newest' is useful.

I see the (new) News column on the home page. The content is right, but it takes too much space on the page. Maybe there could be a box above the "Find us elsewhere" panel that just had news article titles...

Random updates to the wiki pages wouldn't need to be mentioned. But if there's a new topic added, maybe we should publish a corresponding News item to call it out.

>> 	- Can an RSS feed be generated automatically?
> 
> Yes, but see above.

I wonder whether an RSS icon could go in the heading of the News panel...

>> Curating: The current site makes it seem as if all pages are equally important. I feel the urge to do the following:
>> 	- Categorizing the "List of Wiki Pages" for a project (e.g., http://bufferbloat.net/projects/cerowrt/ ) 
>> 		so there's a sense of their importance
> 
> A way to do this would be to update the 'index' wiki pages to be more
> useful (e.g. https://www.bufferbloat.net/projects/cerowrt/wiki/) and
> keep the overview pages as a 'complete' list. However, it's certainly
> also possible to mark some of the pages as 'important' and have those
> show up on top of the list.
> 
> 
>> 	- Discard old/outdated/useless articles
> 
> Yes, this is probably needed. And also fixing broken links (see the
> issue I opened for this on github).

I figure that members of this group can take on sections as the spirit moves them.

>> Workflow:
>> 	- Rules for making posts: Who can make them? How do they get
>> 	published?
> 
> Well, it's a git repo. I figure anyone can do pull requests.

Certainly. This is one of the items that I wanted to be sure we discussed. Dave also mentioned that we might be comfortable giving core authors permission to edit. That would give more direct access for making small tweaks/corrections.

I suspect bigger experiments should be done as PR's (especially if we figure out how to use github.io pages to display the new form of the site from our own repos...)

>> 	- Auto-publish - if it's not already happening, could we
>> 	re-render and publish after a commit?
> 
> Yes, that is certainly possible. For now I figure I'll do it manually
> (it's just re-running a script), but if we get enough activity that that
> becomes a bottleneck, I can certainly set up something automated.

Cool! It looks as if you added code to do that. Is it operational?

>> 	- I see the 404 handler in place, but it doesn't seem to show a
>> 	link to archive.org
> 
> It includes a javascript from archive.org which *should* show a link if
> (and only if) an archived version exists. Haven't verified that it
> actually works.

Ahah! I think this is going to be hard to test - Toke has done too good a job of matching Hugo URLs to the old Redmine URLs... :-)

I did fire up LinkChecker on the site, and it found a few bad URLs. I'll send a separate message about it.

>> 	- Is there a way to do page redirection? It looks as if Toke has done a great job of replicating
>> 		URLs in the new site, but if we see frequent broken
>> 		links, is there a way to redirect a page to its
>> 		equivalent on the new site?
> 
> Yes. This is already used for news items and issues.
> 
>> 	- Should we add a Google search box on 404 page? On every page?
> 
> Not sure what it takes to do this properly, and if it's worth the
> hassle.
> 
>> 	- Would it make sense to review error logs from time to time?
> 
> Possible. Split out the logs to a separate file; can provide it if
> someone wants to go digging. Or I can just parse out a list of 404
> errors from time to time.

I know how to add Google Search, and can take it on after other stuff has settled.

Given the good match between new and old URLs, we might only need a one-time survey of error.log files to see if there's major traffic to a particular "lost" page.

Thanks all!

Rich

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Toke Høiland-Jørgensen]

Rich Brown <richb.hanover@gmail.com> writes:

> Good comments all... Let me preface these responses by saying that I'm
> willing to help out, but don't want to get in the way of the rapid
> progress being made.
>
> Toke: Let us know when you think you're done with major
> reorganizations...

Awesome. I think I'm done with most of the reorganisation stuff now, and
am happy to let other people go nuts :)

> I also want to see if there's consensus about when and how to make
> updates, so that we're not stepping on toes. TL;DR questions:
>
> - Do we only want to use pull requests, or should core authors have
> edit access to github?

Quite happy to give out push access. Only caveat is that there needs to
be an understanding that updates should be tested and if someone breaks
something it is going to break on the live site. However, I figure the
people involved here knows how to deal with that ;)

> - Or do we use the 'draft' facility? (How?)

My thought was that for most things (like news articles, etc), just
writing it up and pushing it 'live' is the right thing to do. But there
may be cases where several people want to explicitly collaborate on,
say, a news article before it is published, in which case the draft
facility can be a way to use the git repo to do this. Using branches
would be another.

>>> 	- Is there a way to display newest entries? I'd like to see a
>>> 	sidebar element showing the 5 newest posts
>> 
>> Yes, certainly. Was thinking of putting a list of newest news items on
>> the front page. However, for the wiki pages it is not necessarily
>> obvious that having a notion of 'newest' is useful.
>
> I see the (new) News column on the home page. The content is right,
> but it takes too much space on the page. Maybe there could be a box
> above the "Find us elsewhere" panel that just had news article
> titles...

Hmm, I do tend to use fairly wide monitors. Can certainly move it if it
takes up too much space. Feel free to play around with it :)

> Random updates to the wiki pages wouldn't need to be mentioned. But if
> there's a new topic added, maybe we should publish a corresponding
> News item to call it out.

I think it's possible to do a 'new items' feed that uses only initial
publication date as the sort key; which would make new wiki pages show
up as well, but not edits of old ones.

>>> 	- Can an RSS feed be generated automatically?
>> 
>> Yes, but see above.
>
> I wonder whether an RSS icon could go in the heading of the News
> panel...

Sure. Just link it to /projects/index.xml - there's already a <link
rel=alternate> in the header.

>> Yes, this is probably needed. And also fixing broken links (see the
>> issue I opened for this on github).
>
> I figure that members of this group can take on sections as the spirit
> moves them.

That was my hope ;)

> I suspect bigger experiments should be done as PR's (especially if we
> figure out how to use github.io pages to display the new form of the
> site from our own repos...)

Yeah. For local experiments, running 'hugo serve' is all you need. For
things that need review, just publishing the files somewhere is all that
is needed. If we end up needing to have a long string of review on some
new feature (say), it is also possible to host automatic builds from
other branches at some other location.

>>> 	- Auto-publish - if it's not already happening, could we
>>> 	re-render and publish after a commit?
>> 
>> Yes, that is certainly possible. For now I figure I'll do it manually
>> (it's just re-running a script), but if we get enough activity that that
>> becomes a bottleneck, I can certainly set up something automated.
>
> Cool! It looks as if you added code to do that. Is it operational?

Yup. Had some more time last night and set it up. It will automatically
pull and build from the master branch whenever it changes. Takes less
than a minute from push to live update :)

>> It includes a javascript from archive.org which *should* show a link if
>> (and only if) an archived version exists. Haven't verified that it
>> actually works.
>
> Ahah! I think this is going to be hard to test - Toke has done too
> good a job of matching Hugo URLs to the old Redmine URLs... :-)
>
> I did fire up LinkChecker on the site, and it found a few bad URLs.
> I'll send a separate message about it.

Saw that in the logs. I am including a filtered error log below, listing
the wiki URLs that seem to be missing.

>> Possible. Split out the logs to a separate file; can provide it if
>> someone wants to go digging. Or I can just parse out a list of 404
>> errors from time to time.
>
> I know how to add Google Search, and can take it on after other stuff
> has settled.

Awesome.

> Given the good match between new and old URLs, we might only need a
> one-time survey of error.log files to see if there's major traffic to
> a particular "lost" page.

See list below.

-Toke

$ grep 404 /var/log/nginx/bufferbloat.access.log | awk '{print $7}' | sort | uniq | grep wiki | egrep -v 'annotate|history|diff|edit|png|version=|jpe?g'    
/login?back_url=http://www.bufferbloat.net/projects/wisp6/wiki/Wiki
/projects/bismark/wiki
/projects/bismark/wiki/Capetown
/projects/bismark/wiki/Mac_Instructions
/projects/bismark/wiki/Wiki
/projects/bismark/wiki/Wndr3700v2
/projects/bloat/wiki/ARP
/projects/bloat/wiki/bql_enabled_drivers
/projects/bloat/wiki/Bufferbloat_and_Freeswitch_Conference_Call_March_9/needthelink
/projects/bloat/wiki/Daddy_why_is_my_Internet_Slow_Today
/projects/bloat/wiki/Dark_Buffers
/projects/bloat/wiki/Dogfood_principle
/projects/bloat/wiki/experiment-fun_with_your_switch
/projects/bloat/wiki/index
/projects/bloat/wiki/introduction
/projects/bloat/wiki/mitigations_and_solutions_for_broadband
/projects/bloat/wiki/Mitigations_and_solutions_for_Home_Gateways
/projects/bloat/wiki/Papers/href=
/projects/bloat/wiki/technicalintro
/projects/bloat/wiki/todo
/projects/bloat/wiki/VoIP
/projects/bloat/wiki/VOIP/m2e-delay.PNG
/projects/bloat/wiki/What_(bad)_stuff_happens_on_a_congested_network
/projects/bloat/wiki/Wiki
/projects/cerowrt/wiki/BloatLab_1/
/projects/cerowrt/wiki/BloatLab_1/attached_t.svg
/projects/cerowrt/wiki/cerowrt_flashing_instructions
/projects/cerowrt/wiki/CeroWrt_flashing_instructions
/projects/cerowrt/wiki/evaluating_qos_behavior
/projects/cerowrt/wiki/Experiment_-_QoS
/projects/cerowrt/wiki/fq_codel_on_wireless
/projects/cerowrt/wiki/Installation_guide
/projects/cerowrt/wiki/ipv6_tunnel
/projects/cerowrt/wiki/Newer_versions_of_fq_codel
/projects/cerowrt/wiki/Paris
/projects/cerowrt/wiki/Pre-33_Releases
/projects/cerowrt/wiki/Setting_Up_SQM_for_CeroWrt_310
/projects/cerowrt/wiki/uftp
/projects/cerowrt/wiki/Wiki
/projects/cerowrt/wiki/wondershaper_must_die
/projects/codel/wiki/best_practices_for_benchmarking_codel_and_fq_codel
/projects/codel/wiki/Best_Practices_for_Benchmarking_CoDel_and_FQ_CoDel
/projects/codel/wiki/howto
/projects/codel/wiki/index
/projects/codel/wiki/rrul_test_suite
/projects/codel/wiki/Wiki
/projects/codel/wiki/Wiki/
/projects/iscwrt/wiki/
/projects/iscwrt/wiki/Wiki
/projects/make-wifi-fast/wiki/Wiki
/projects/projects/bloat/wiki/BQL_enabled_drivers/
/projects/projects/cerowrt/wiki/CeroWrtScripts/
/projects/projects/cerowrt/wiki/Debugging_CeroWrt/
/projects/projects/cerowrt/wiki/Enable_ECN/
/projects/projects/cerowrt/wiki/Quick_Test_for_Bufferbloat/
/projects/projects/cerowrt/wiki/SQM/
/projects/projects/cerowrt/wiki/Wondershaper_Must_Die/
/projects/projects/codel/wiki/Best_practices_for_benchmarking_Codel_and_FQ_Codel/
/projects/projects/codel/wiki/Bobbie/
/projects/projects/codel/wiki/Cake/
/projects/uberwrt/wiki
/projects/uberwrt/wiki/Experimental_patches
/projects/uberwrt/wiki/Hardware_evaluation

[Bloat] Bufferbloat.net & LinkChecker

[Rich Brown]

For your site consistency happiness...

I am fond of the LinkChecker program that spiders a site and checks for broken links, questionable HTML, etc. It's at: http://wummel.github.io/linkchecker/ It has a GUI for Windows and Linux (sadly, the GUI doesn't build on OSX, but the CLI works) 

Fire off the CLI to get a CSV report (with ";" separators):

linkchecker http://bufferbloat.net > ~/bufferbloat-errs.csv -o csv

I have pasted the resulting CSV file at: http://pastebin.com/XaFUjKEL

Rich

Re: [Bloat] Bufferbloat.net & LinkChecker

[Toke Høiland-Jørgensen]

Rich Brown <richb.hanover@gmail.com> writes:

> I have pasted the resulting CSV file at: http://pastebin.com/XaFUjKEL

Cool. Now someone just needs to go fix all those ;)


I note that I have other stuff to attend to, so I am not expecting to do
more work on the site for the time being. So don't be shy, go fix stuff
(that goes for everyone, not you in particular). I'll be happy to review
pull requests and/or assist with explaining some of the ways I've had to
wrangle Hugo to get it to do stuff the way I wanted it to, as needed.

-Toke

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Dave Taht]

On Sun, Jun 12, 2016 at 10:49 AM, Rich Brown <richb.hanover@gmail.com> wrote:
> To all the Bufferbloaters out there...
>
> Thanks again to Toke for all this good work. It's terrific to see the pages resurrected, and to have the www.bufferbloat.net site living again.

And it's fast - even half a planet away. I am still looking around at
various cdn technologies....

> I've been thinking of the next steps - to make all this work really useful. I see there are three important issues:
>
> Organizing: Thinking about the best way to display the current information:
>         - Should Cake be listed as a top-level project?

"smart queue management" could, with cake mentioned prominently. Put
in an overview for all the places fq_codel has made it to
(edgerouters, sophos, BSD, openwrt, dd-wrt, (?) - what we know of
certain commercial products when they did it right (adaptive qos), or
didn't quite (streamboost).

"cake" is such a terrible name googlejuicewise. I do periodically
google for "cake shaper" and giggle at the results... which reminds me
that somehow, someday, a logo(s) for the project(s) would be nice.

>         - Is there a way to display newest entries? I'd like to see a sidebar element showing the 5 newest posts

Jim and I had hoped to build a "blog planet" one day that federated
all the #bufferbloat tags from various sources. Since this is a manual
process currently (I still have a daily google query for it), perhaps
automating that with javascript to (somehow) grab the last twitter and
g+ mentions, and a few key blogs, or building it on the fly on a
build?

>         - Can an RSS feed be generated automatically?

Yes.

> Curating: The current site makes it seem as if all pages are equally important. I feel the urge to do the following:
>         - Categorizing the "List of Wiki Pages" for a project (e.g., http://bufferbloat.net/projects/cerowrt/ )
>                 so there's a sense of their importance

Um, I figured this would be just autogenerated from everything published.

There are ways to mark the metadata as to priority.

>         - Discard old/outdated/useless articles
or those that need work, remark to be draft: yes

> Workflow:
>         - Rules for making posts: Who can make them? How do they get published?

There is a difference between a "blog post/news item" and a wiki page.

One reason why I intend to keep blog.cerowrt.org separate, is there I
am trying to build upon the methods I was using with the old "deBloat"
repository, "having a public lab notebook, chock full of opinions,
wild theories, and mis-interpretations" - that this time, is
replicated with all the data that backs (or doesn't back) all that up.

I am still quite remorseful that all my private notes and data from
the first 2 years of the effort was lost when my lab machines got
stolen, there was plenty of stuff I did not publish there that I wish
I could refer to now were I still trying for a PHD. The deBloat repo
and various presentations are the only thing that survived from that
era. Never again! but that means that that blog repo is going to get
very, very big, and I'd prefer these sites to be primarily text and
graphics, and thus smaller.

I note that in blog.cerowrt.org I am preserving page authorship (and
do invite other posters!).

Toke has this stuff copyright "the bufferbloat community", which is a
meaningless abstraction legally. the site overall was basically
creative commons. I don't think anybody here cares.

>         - Auto-publish - if it's not already happening, could we re-render and publish after a commit?
>         - I see the 404 handler in place, but it doesn't seem to show a link to archive.org

My plan for 404's long term was to have a page explaining this was a
volunteer project, encouraging them to fix their own bloat, and
contribute either work or cash. I thought about using a canned copy of
this:

http://www.dslreports.com/speedtest/results/bufferbloat?up=1

with the popular "you are here/you could be here" annotation.

I wish there was some way to sample the tcp stats on the page load
itself to measure the bloat the user experienced....

>         - Is there a way to do page redirection? It looks as if Toke has done a great job of replicating
>                 URLs in the new site, but if we see frequent broken links, is there a way to redirect a page to its equivalent on the new site?
>         - Should we add a Google search box on 404 page? On every page?

I have missed having search, especially on the bug tracker. It was one
of those bugs I thought I'd fix in "in a few weeks", 5 years ago.

As the plan is to move to github's bug tracker, that's solved. As for
indexing the site, well, I keep hoping that various non-google search
technologies (like Xapian) had progressed in usability.

Is there a way to embed something that uses no javascript?

>         - Would it make sense to review error logs from time to time?

Sure, although we could add the linkchecker to people's local builds.

Definately agree it would be good to track what people have been using
as inbound links, on plain access also. People kept linking to the
oddest parts of the site...



>
> I'd love to have your thoughts. Thanks.
>
> Rich



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Toke Høiland-Jørgensen]

Dave Taht <dave.taht@gmail.com> writes:

> On Sun, Jun 12, 2016 at 10:49 AM, Rich Brown <richb.hanover@gmail.com> wrote:
>> To all the Bufferbloaters out there...
>>
>> Thanks again to Toke for all this good work. It's terrific to see the pages
>> resurrected, and to have the www.bufferbloat.net site living again.
>
> And it's fast - even half a planet away. I am still looking around at
> various cdn technologies....

I played around with using cloudfare through amazon for my blog at some
point; for static pages you can get it to pull from an S3 bucket.
Syncing to that is fairly straight forward.

I gave up on the effort at the time because I botched the SSL cert setup
(cloudfare only accepts up to 2048 bits, and this was pre-letsencrypt,
so getting a new one once issued was non-trivial), and because the
propagation delay for changes was too high.

We can certainly stick it in a CDN if you find something that works
well, but for now at least I'd call that a premature optimisation.

I did configure the web server to do most of the obvious speed
optimisations (for instance, everything will be statically compressed on
build, so the server basically just has to do sendfile() on the right
file on request).

https://developers.google.com/speed/pagespeed/insights/?url=https://www.bufferbloat.net/projects/


> Definately agree it would be good to track what people have been using
> as inbound links, on plain access also. People kept linking to the
> oddest parts of the site...

I can have the server generate awstats reports for the site
(http://www.awstats.org/) - these include the most common inbound
referrers and most common 404 pages. Not sure if they should be public,
though. Don't *think* they usually contain sensitive content (given that
the whole site is public, anyway). But I could be wrong; thoughts?

-Toke

Re: [Bloat] Bufferbloat.net - Organizing, curating, and workflow

[Dave Taht]

as for "Search in the age of slack", I totally missed that if we are
going to do anything dynamic-ish and automagic for the front pages,
that pulling a few entries from the mailing lists would probably help.
Similarly the search engine should also incorporate
lists.bufferbloat.net content.

At the distributed web conference last week, I was stunned by a
sea-change in interactions, everybody posted their twitter handles,
instead of email addresses, and were interacting over slack (rather
than irc) or federating via matrix.org. I do not know to what extent
the old and new can be federated together....

The new front page looks good on a handheld but is not particularly
useful. I'd move make-wifi-fast to the top and make it a link, in
particular.

The opening day of that conference is up on youtube with talks by
mitchel baker, vint cerf, tim berners-lee, and cory doctorow, all
worthwhile...

https://www.youtube.com/watch?v=Yth7O6yeZRE

Doctorow's talk (about 5 hours in) was particularly inspiring. I wish
I'd got a lightning talk in, as "TCP 101" and the effects of
bufferbloat on edge-distributed content seems to have been missed by
the new guard.

http://blog.dshr.org/2016/04/brewster-kahles-distributed-web-proposal.html

In terms of other bleeding edge distribution concepts there's things like ipfs.

Re: [Bloat] bufferbloat.net is sorely missed/Now it's back!

[Kevin Darbyshire-Bryant]

On 11/06/16 21:05, Toke Høiland-Jørgensen wrote:
> Rich Brown <richb.hanover@gmail.com> writes:
>
>> The new site just went live here in the shrubs of New Hampshire, USA.
>> (It wasn't working ~10 minutes ago.)
> Awesome! The letsencrypt servers seem to have picked up the change too,
> so there's now a cert in place and non-https redirects to serve via
> https.
>

Thank you to all involved in getting that sorted out.  I don't know how 
you did it, it's magic to me, fantastic result and very much appreciated.

Cheers,

Kevin

Re: [Bloat] bufferbloat.net is sorely missed/Now it's back!

[Toke Høiland-Jørgensen]

Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk> writes:

> Thank you to all involved in getting that sorted out. I don't know how
> you did it, it's magic to me, fantastic result and very much
> appreciated.

Thanks! :)

The 'how' is simply: Python! Available here for your perusal:
https://github.com/tohojo/bufferbloat-net/tree/master/export

Obligatory XKCD: https://xkcd.com/353/

-Toke