Green vs. white markings in the HCL

[Joanna Rutkowska]

"
Green - The reporter has clearly stated that all components and features
are working correctly.

White - The reporter has sent an HCL Info output which indicates
everything is OK without any comments.
"

So, can we please get rid of this, somehow superficial, distinction
between the green and white markings in the HCL and just use green
everywhere? If not for simplicity and common sense, then at least for
"marketing purposes"? :)

I guess we could do that over the next days when HCL will be updated to
reflect R2B3 compatibility (which, BTW, should be a subset of R2B2
because it will be released with 3 kernels, for both installer and
installed system, one of which being 3.7 as in R2B2)

Thanks,
joanna.

[7v5w7go9ub0o]

As more folks use Qubes, the distinction would become more important - a
well-intended, but less-than-expert contributor may incorrectly believe
and report that everything is o.k.

Suggest that everyone reporting to the HCL be thanked and strongly
encouraged to use the script, and that you drop the green.

[Joanna Rutkowska]

Let's not make things more complicated than they are.

Besides, your argument doesn't hold water -- a less-than-expert
contributor might very well send lots of endorsements that will turn out
to be false. There is no way to verify contributed reports, and hence
the distinction in the HCL between the machines tested by Qubes Core
Devels vs. the Rest of the World.

joanna.

[7v5w7go9ub0o]

On 12/06/13 18:46, Joanna Rutkowska wrote:
> On 12/06/13 22:20, 7v5w7go9ub0o wrote:
>> On 12/06/13 15:43, Joanna Rutkowska wrote:
>>> " Green - The reporter has clearly stated that all components
>>> and features are working correctly.
>>>
>>> White - The reporter has sent an HCL Info output which indicates
>>> everything is OK without any comments. "
>>>
>>> So, can we please get rid of this, somehow superficial,
>>> distinction between the green and white markings in the HCL and
>>> just use green everywhere? If not for simplicity and common
>>> sense, then at least for "marketing purposes"? :)
>>>
>>> I guess we could do that over the next days when HCL will be
>>> updated to reflect R2B3 compatibility (which, BTW, should be a
>>> subset of R2B2 because it will be released with 3 kernels, for
>>> both installer and installed system, one of which being 3.7 as in
>>> R2B2)
>>
>> As more folks use Qubes, the distinction would become more
>> important - a well-intended, but less-than-expert contributor may
>> incorrectly believe and report that everything is o.k.
>>
>> Suggest that everyone reporting to the HCL be thanked and strongly
>> encouraged to use the script, and that you drop the green.
>>
>>
>
> Let's not make things more complicated than they are.

Agreed.

Require scripts for "everything is OK" ratings.

>
> Besides, your argument doesn't hold water -- a less-than-expert
> contributor might very well send lots of endorsements that will turn
> out to be false. There is no way to verify contributed reports, and
> hence the distinction in the HCL between the machines tested by
> Qubes Core Devels vs. the Rest of the World.

1. If you accept scripts only, you don't have to verify "everything is
OK" reports.

2. The primary users of the HCL will be prospective buyers wanting to
upgrade their boxes. They will certainly consider only "everything is
OK" hardware - so that high rating had better be solid.

[Joanna Rutkowska]

Another wrong assumption. Everybody can spoof a HCL, even if it could
use hundreds of scripts to check all the functionality.

Besides it is not possible to write scripts to check lots of things.

Unless... you would like to contribute some? :)

> 2. The primary users of the HCL will be prospective buyers wanting to
> upgrade their boxes. They will certainly consider only "everything is
> OK" hardware - so that high rating had better be solid.
>

Not possible for it to be solid. Qubes HCL is a best effort thing. Just
like the Internet.

joanna.

[Zrubecz Laszlo]

On 6 December 2013 21:43, Joanna Rutkowska

<joa...@invisiblethingslab.com> wrote:
> So, can we please get rid of this, somehow superficial, distinction
> between the green and white markings in the HCL and just use green
> everywhere? If not for simplicity and common sense, then at least for
> "marketing purposes"? :)

Well... sure we can do.
Actually the hardest thing to decide if it the standard devices are
really working under linux or not.
While it is really not Qubes specific, but a basic requirement.
(just a side note that even the biggest Linux distros are still not
able to maintain such a HCL)

> I guess we could do that over the next days when HCL will be updated to
> reflect R2B3 compatibility (which, BTW, should be a subset of R2B2
> because it will be released with 3 kernels, for both installer and
> installed system, one of which being 3.7 as in R2B2)

Then we also have to decide how to handle the whole HCL data.
As far as I remember we were all agreed about to keep the latest
stable, and the latest Beta release, right?
I suggest that we should archive this state, and start a new table by
clearing the R2B2 reports.


--
Zrubi

[Axon]

Joanna Rutkowska:

It should be sufficient to note, on the HCL webpage, that the reports
are generated and submitted by users (and that they may, therefore, be
inaccurate). That's the best the Qubes team can do, since the team can't
obtain and test every piece of hardware (allegedly) tested by Qubes
users. (I've added such a note at the top of the page.)

I'm in favor of collapsing the green/white distinction.

[Zrubecz Laszlo]

On 7 December 2013 12:09, Zrubecz Laszlo <ma...@zrubi.hu> wrote:

> I suggest that we should archive this state, and start a new table by
> clearing the R2B2 reports.

Just started to clear the actual HCL.

And made a backup of the R2B2 state:
https://wiki.qubes-os.org/trac/wiki/HCL-R2B2

If it is not needed anymore, can be deleted.
(I have a local copy to work from)


--
Zrubi

[Joanna Rutkowska]

You can start by marking my T420s as all green :)

j.

[Zrubecz Laszlo]

On 11 December 2013 10:16, Joanna Rutkowska

<joa...@invisiblethingslab.com> wrote:
> On 12/11/13 09:19, Zrubecz Laszlo wrote:

>> Just started to clear the actual HCL.

> You can start by marking my T420s as all green :)

Done.


I decided to keep the "surely not working" hardwares because these
should be avoided.


--
Zrubi

[Joanna Rutkowska]

BTW, I think you cold add an explanation that if there is a white entry
without anything written in it, it means simply that this particular
system hasn't been tested yet.

BTW, there is no line separating Sony Vaio and Dell in the table.

Marek can you confirm "all green" for your Dell and perhaps Samsung as well?

joanna.

[Joanna Rutkowska]

Aha, Zrubi, and for Vaio Z 2010 -- it's the same as for R1 -- i.e. all
works fine except for the tinkering needed as described in the Remarks.

j.

[Joanna Rutkowska]

On 12/11/13 10:27, Zrubecz Laszlo wrote:

But why do you assume those platforms are still all-red for R2B3? We use
different kernels there...

j.

[Zrubecz Laszlo]

On 11 December 2013 10:36, Joanna Rutkowska

<joa...@invisiblethingslab.com> wrote:
> But why do you assume those platforms are still all-red for R2B3? We use
> different kernels there...

All those kernels have been available before...


--
Zrubi

[Marek Marczykowski-Górecki]

Yes, all green on my Dell. Also on Thinkpad T61 and Samsung (with exception
for VT-d on both of them).

--
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?

[Zrubecz Laszlo]

On 14 December 2013 05:39, Marek Marczykowski-Górecki

<marm...@invisiblethingslab.com> wrote:

> Yes, all green on my Dell. Also on Thinkpad T61 and Samsung (with exception
> for VT-d on both of them).

Dear "Qubes core developers",

Can you please RUN the HCL report script ony your machine(s), and
actually send the output please? :)

While I beleive that your reports are accurate, the HCL is missing a
lot of useful information abot the "Qubes core developers" reported
machines.

Beside that, I think we should replace the OK sign from "Standard
Features" to the best running kernel version.
(Thats why I made the change to include the running kernel version as well)


Many thanks.

--
Zrubi

[Joanna Rutkowska]

On 12/14/13 05:39, Marek Marczykowski-Górecki wrote:
> On 11.12.2013 10:32, Joanna Rutkowska wrote:
>> On 12/11/13 10:27, Zrubecz Laszlo wrote:
>>> On 11 December 2013 10:16, Joanna Rutkowska
>>> <joa...@invisiblethingslab.com> wrote:
>>>> On 12/11/13 09:19, Zrubecz Laszlo wrote:
>>>
>>>>> Just started to clear the actual HCL.
>>>
>>>> You can start by marking my T420s as all green :)
>>>
>>> Done.
>>>
>>>
>>> I decided to keep the "surely not working" hardwares because these
>>> should be avoided.
>>>
>>>
>> BTW, I think you cold add an explanation that if there is a white entry
>> without anything written in it, it means simply that this particular
>> system hasn't been tested yet.
>>
>> BTW, there is no line separating Sony Vaio and Dell in the table.
>>
>> Marek can you confirm "all green" for your Dell and perhaps Samsung as well?
>
> Yes, all green on my Dell. Also on Thinkpad T61 and Samsung (with exception
> for VT-d on both of them).
>

If they don't support VT-d, then it is yellow, not green.

j.

[7v5w7go9ub0o]

On 12/14/13 03:40, Zrubecz Laszlo wrote:
> On 14 December 2013 05:39, Marek Marczykowski-G�recki

> <marm...@invisiblethingslab.com> wrote:
>
>> Yes, all green on my Dell. Also on Thinkpad T61 and Samsung (with
>> exception for VT-d on both of them).
>
> Dear "Qubes core developers",
>
> Can you please RUN the HCL report script ony your machine(s), and
> actually send the output please? :)
>
> While I beleive that your reports are accurate, the HCL is missing a
> lot of useful information abot the "Qubes core developers" reported
> machines.
>
> Beside that, I think we should replace the OK sign from "Standard
> Features" to the best running kernel version. (Thats why I made the
> change to include the running kernel version as well)

Good move.

Also, is it now time to add an AEM column?

[Marek Marczykowski-Górecki]

On 14.12.2013 09:40, Zrubecz Laszlo wrote:
> On 14 December 2013 05:39, Marek Marczykowski-Górecki
> <marm...@invisiblethingslab.com> wrote:
>
>> Yes, all green on my Dell. Also on Thinkpad T61 and Samsung (with exception
>> for VT-d on both of them).
>
> Dear "Qubes core developers",
>
> Can you please RUN the HCL report script ony your machine(s), and
> actually send the output please? :)
>
> While I beleive that your reports are accurate, the HCL is missing a
> lot of useful information abot the "Qubes core developers" reported
> machines.

Attached HCL info for two of them. Thinkpad one will send later.

[Hakisho Nukama]

Can you report, if every kernel is working properly?

Thanks, added.

[Marek Marczykowski-Górecki]

Yes, on both machines.

> Thanks, added.