Detailed technical computer model number and hardware specs should be included on customer facing certification site

Bug #664088 reported by komputes
24
This bug affects 3 people
Affects Status Importance Assigned to Milestone
certify-planning
Fix Released
Undecided
David Murphy
certify-web
Fix Released
Medium
David Murphy

Bug Description

We should use a more detailed/technical model number on http://webapps.ubuntu.com/certification. We should also include hardware component information (and perhaps even an FAQ on how to get this info from a machine using dmidecode) to avoid confusion between two machines with the same human name, but different technical model numbers and different components. Two machines can have the same name, but can be quite different under the hood.

Example/Use Case:
Placing the statement "The Acer Aspire One netbook has been awarded the status of Certified on Ubuntu PC (x86)" [1] on webapps.ubuntu.com/certification without specifying hardware details gets buyers of this hardware upset as this list is used by many as an informational pre-purchase resource. In this case we certified a AOA110 and the customer purchased a AOD260 which has an unsupported card reader. Using our website, or Acer's website there was no way the buyer could have known this. We currently give the impression that all Aspire One models are certified to run Ubuntu.

Fault lies in the marketing of the device, since the manufacturer sold multiple machine configurations under the same name. We can improve and avoid this issue by including detailed technical information on the publicly available certification site/webapp.

[1] http://webapps.ubuntu.com/certification/hardware/200905-2911/#1004-lts

Tags: cer-1535
Changed in certify-planning:
milestone: none → may2011
Changed in certify-web:
assignee: nobody → David Murphy (schwuk)
Changed in certify-planning:
assignee: nobody → David Murphy (schwuk)
Changed in certify-planning:
milestone: may2011 → none
David Murphy (schwuk)
Changed in certify-web:
status: New → Confirmed
importance: Undecided → Medium
milestone: none → 2011.13
David Murphy (schwuk)
visibility: private → public
Revision history for this message
Jarl (jarl-dk) wrote :

I will comment on the tittle of this bug:
"Detailed technical computer model number and hardware specs should be included on customer facing certification site"

I would say for the certification list to be usefull as a buying guide, the list must include SKU number, if the list contains SKU number, then detailed HW description should be unnecessary and maybe even confusing if it does not match the specs on the manufacturers site.

Another thing....

Why is bug #650550 marked duplicate of this one, I believed that the convention was that the first bug was the main bug and all later bugs (like this one) was duplicates...

Jarl

Revision history for this message
Alan Pope 🍺🐧🐱 🦄 (popey) wrote :

@Jarl sometimes a later bug is a dupe of an earlier one, sometimes vice versa. It doesn't really matter so long as the necessary data is captured.

Revision history for this message
Jarl (jarl-dk) wrote :

@Alan, you may be right from a bug-fixers point of view. But things look different from a bug-reporters point of view. I consider it quite dis-respectful to the bug reporter. Quite some time and effor is spent to make good valuable bug reports (here among searching for existing identical bugs), if this effort is not respected bug repporters wil most likely not put so much effort into the process in the future since the effort (searching for existing identical bugs) is (apparently) not valued and respected anyway.

Marking a bug as duplicate sends two messages to the bug reporter:
1) Your effor tin reporting the bug has been waste of your time, since another person have reported the same issue.
2) You have waste the time of bug-fixers (by not searching for identical bug) because they have spent time on identifying it as a duplicate.

Tell me. Why would I ever want to search for previously reported bugs of identical issue if nobody (no other bug reporters) else does, and nobody (no bug-fixers) really cares if I do it or not?

Jarl

Revision history for this message
Robert Collins (lifeless) wrote :

@Jarl your analysis of the impact and time use of duplicates being filed is incorrect.

Many times a different report with the same root cause will expose information the first report missed. Sometimes many reports are needed to triangulate and identify the root cause. So filing a duplicate is not /necessarily/ a waste of time.

Secondly finding duplicates is often much harder for a bug filer than a developer of the package/product - because bug filers don't know the architecture.

Lastly, particularly in complex components like the Linux kernel, developers *do not want* users to search for duplicates: there are so many symptoms that look similar, that it causes a lot of harm when user A comments on a bug filed by user B.

All you need to do to be a good bug filer is:
 - follow the bug filing wizard
 - if it offers a duplicate up, and you think its the same, click on affects me too
 - if you get asked to provide more information, do so

Revision history for this message
Jarl (jarl-dk) wrote :

@Robert. Thanks! I agree, I especially see the point in the case of large software component such as the Linux kernel. I have seen several cases where bug-filers comments have confused developers more than helped due to laziness of not reporting a separate bug. I also see the point in apport and bug filing wizard suffers in detecting duplicates (I wonder why bug #650550 didn't show up for this one).

Nevertheless I have also seen (apart from this one) seen examples where bug-filers have felt offended by their bugs being marked duplicates of later bugs.

Though, an interesting discussion, it should take place somewhere else... I will end here with the open question how does duplication marks affect the bug-filers karma? For the bug-filer of the duplicate bug? and for the bug-filer of the main bug?

Jarl

Revision history for this message
David Murphy (schwuk) wrote :

The public certification site now contains details of devices in a given model.

Changed in certify-web:
status: Confirmed → Fix Released
Changed in certify-planning:
status: New → Fix Released
Matias Piipari (mz2)
tags: added: cer-1535
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.