Call for VOTE: OGF GLUE2 Enumerations procedures and best practices
GLUE WG:
From our last GLUE WG teleconference June 3, 2014: Action 2: GLUE WG to review Florido's revised Enumerations document Action 2: At the next GLUE WG call we will vote on the updated document
We will be taking a formal vote tomorrow, June 17, to accept the "OGF GLUE2 Enumerations procedures and best practices" as our practice. Please review the document here: odt: https://redmine.ogf.org/dmsf_files/13260?download= pdf: https://redmine.ogf.org/dmsf_files/13261?download= If you have opinion on this best practices document please reply with: APPROVE - if you support this best practice DISAPPROVE - if you object, please clearly summarize your objections so that we can discuss them If you approve but have suggested please reply with your approval and list your suggestions. Regards, JP
Dear Glue WG, below are some comments to the document. Best, Andre. - page 3: Exisiting -> Existing - page 3: 'organization_name' is marked as RECOMMENDED (not MUST), but later on is part of the 'minimum requirement' -- that seems to be a contradiction? - page 4: 'should be RECOMMENDED' -> 'is RECOMMENDED' - page 4: 'organization_name' and 'interface_name' are MANDATORY, but '<organization_name>.<interface_name>' is 'RECOMMENDED'. Are there real alternatives to the latter or should that be a 'MUST' ? - page 5: 'org.ogf.glue.* SHOULD be used if there are issues with domain names..' - I am not sure what that means. What issues? Also, there are two points at the end of sentence. - 'If a project ends and the product has an orphaned reverse domain name, the name can be kept.' - this reads like the orphaned name is the one to be kept -- is that correct? - line 119: should -> SHOULD - line 128: should not -> SHOULD NOT ; bu -> by - line 130: RECOMMENDED NOT -> NOT RECOMMENDED - line 133: cannot be -> MUST NOT be - line 149: 'If the values are accepted, the Status field SHOULD contain Recommended.' -> isn't that a MUST? If not, what other value could it be? - line 151: 'As soon as possible' -> might want to replace with 'at the next group meeting or call'? ASAP does not mean much ;) - a few documents are referenced in the text, but proper references/links are at the end of the document are missing. On Mon, Jun 16, 2014 at 4:41 PM, Navarro, John-Paul F. <navarro@mcs.anl.gov> wrote:
GLUE WG:
From our last GLUE WG teleconference June 3, 2014:
Action 2: GLUE WG to review Florido's revised Enumerations document Action 2: At the next GLUE WG call we will vote on the updated document
We will be taking a formal vote tomorrow, June 17, to accept the "OGF GLUE2 Enumerations procedures and best practices" as our practice.
Please review the document here: odt: https://redmine.ogf.org/dmsf_files/13260?download= pdf: https://redmine.ogf.org/dmsf_files/13261?download=
If you have opinion on this best practices document please reply with:
APPROVE - if you support this best practice DISAPPROVE - if you object, please clearly summarize your objections so that we can discuss them
If you approve but have suggested please reply with your approval and list your suggestions.
Regards,
JP _______________________________________________ glue-wg mailing list glue-wg@ogf.org https://www.ogf.org/mailman/listinfo/glue-wg
-- It was a sad and disappointing day when I discovered that my Universal Remote Control did not, in fact, control the Universe. (Not even remotely.)
Thanks Andre for the comments! I'll try to fix the trivial ones by the beginning of the meeting, but I can't promise. About some of them, here's my own comments: On 2014-06-17 10:25, Andre Merzky wrote:
Dear Glue WG,
below are some comments to the document.
Best, Andre.
- page 3: Exisiting -> Existing - page 3: 'organization_name' is marked as RECOMMENDED (not MUST), but later on is part of the 'minimum requirement' -- that seems to be a contradiction?
I actually don't like RECOMMENDED there, but as a matter of fact we have names with no organization prefix. My choice would be to keep old names, but for new names that have no organization prefix we should force "org.ogf.glue." as an organization prefix to keep consistency. I think Stephen was against it, AFAIR. Stephen?
- page 4: 'organization_name' and 'interface_name' are MANDATORY, but '<organization_name>.<interface_name>' is 'RECOMMENDED'. Are there real alternatives to the latter or should that be a 'MUST' ?
See previous comment. It also says MANDATORY for NEW names. Maybe I should stress the NEW?
- page 5: 'org.ogf.glue.* SHOULD be used if there are issues with domain names..' - I am not sure what that means. What issues?
for example a domain name is bought by somebody else who is alien to the initial project. Or simply the domain name does not exist. I just received a request today with domain names that exist but are totally unrelated to the described services. Listing all the issues would not make sense. Is it a bit clearer now? Do you have better suggestions for rephrasing it?
Also, there are two points at the end of sentence. - 'If a project ends and the product has an orphaned reverse domain name, the name can be kept.' - this reads like the orphaned name is the one to be kept -- is that correct?
It is correct.
[...] - line 149: 'If the values are accepted, the Status field SHOULD contain Recommended.' -> isn't that a MUST? If not, what other value could it be?
I think that line should be changed -- in fact it could be any of the other values as well. I think the best is to remove that line and just say that the group should assign one of the values defined in Section 4.2 -- without a value the string is still floating in a DRAFT state.
- line 151: 'As soon as possible' -> might want to replace with 'at the next group meeting or call'? ASAP does not mean much ;)
Sounds good for me!
- a few documents are referenced in the text, but proper references/links are at the end of the document are missing.
True, will add them. Regards, Florido
On Mon, Jun 16, 2014 at 4:41 PM, Navarro, John-Paul F. <navarro@mcs.anl.gov> wrote:
GLUE WG:
From our last GLUE WG teleconference June 3, 2014:
Action 2: GLUE WG to review Florido's revised Enumerations document Action 2: At the next GLUE WG call we will vote on the updated document
We will be taking a formal vote tomorrow, June 17, to accept the "OGF GLUE2 Enumerations procedures and best practices" as our practice.
Please review the document here: odt: https://redmine.ogf.org/dmsf_files/13260?download= pdf: https://redmine.ogf.org/dmsf_files/13261?download=
If you have opinion on this best practices document please reply with:
APPROVE - if you support this best practice DISAPPROVE - if you object, please clearly summarize your objections so that we can discuss them
If you approve but have suggested please reply with your approval and list your suggestions.
Regards,
JP _______________________________________________ glue-wg mailing list glue-wg@ogf.org https://www.ogf.org/mailman/listinfo/glue-wg
-- ================================================== Florido Paganelli ARC Middleware Developer - NorduGrid Collaboration System Administrator Lund University Department of Physics Division of Particle Physics BOX118 221 00 Lund Office Location: Fysikum, Hus B, Rum B313 Office Tel: 046-2220272 Email: florido.paganelli@REMOVE_THIShep.lu.se Homepage: http://www.hep.lu.se/staff/paganelli ==================================================
Thanks for the quick reply! Comments/replies inlined below. Cheers, Andre. On Tue, Jun 17, 2014 at 11:03 AM, Florido Paganelli <florido.paganelli@hep.lu.se> wrote:
- page 4: 'organization_name' and 'interface_name' are MANDATORY, but '<organization_name>.<interface_name>' is 'RECOMMENDED'. Are there real alternatives to the latter or should that be a 'MUST' ?
See previous comment. It also says MANDATORY for NEW names. Maybe I should stress the NEW?
Ah, I overlooked the 'new'. The statements are correct then IMHO. But yes, emphasizing may make it clearer.
- page 5: 'org.ogf.glue.* SHOULD be used if there are issues with domain names..' - I am not sure what that means. What issues?
for example a domain name is bought by somebody else who is alien to the initial project. Or simply the domain name does not exist.
I just received a request today with domain names that exist but are totally unrelated to the described services.
Listing all the issues would not make sense. Is it a bit clearer now? Do you have better suggestions for rephrasing it?
Maybe instead 'if there are issues' use something like 'if a domain name does not exist or is not applicable'? Either way though, thanks for the clarification, makes sense. Thanks, Andre.
Regards, Florido
On Mon, Jun 16, 2014 at 4:41 PM, Navarro, John-Paul F. <navarro@mcs.anl.gov> wrote:
GLUE WG:
From our last GLUE WG teleconference June 3, 2014:
Action 2: GLUE WG to review Florido's revised Enumerations document Action 2: At the next GLUE WG call we will vote on the updated document
We will be taking a formal vote tomorrow, June 17, to accept the "OGF GLUE2 Enumerations procedures and best practices" as our practice.
Please review the document here: odt: https://redmine.ogf.org/dmsf_files/13260?download= pdf: https://redmine.ogf.org/dmsf_files/13261?download=
If you have opinion on this best practices document please reply with:
APPROVE - if you support this best practice DISAPPROVE - if you object, please clearly summarize your objections so that we can discuss them
If you approve but have suggested please reply with your approval and list your suggestions.
Regards,
JP _______________________________________________ glue-wg mailing list glue-wg@ogf.org https://www.ogf.org/mailman/listinfo/glue-wg
-- ================================================== Florido Paganelli ARC Middleware Developer - NorduGrid Collaboration System Administrator Lund University Department of Physics Division of Particle Physics BOX118 221 00 Lund Office Location: Fysikum, Hus B, Rum B313 Office Tel: 046-2220272 Email: florido.paganelli@REMOVE_THIShep.lu.se Homepage: http://www.hep.lu.se/staff/paganelli ================================================== _______________________________________________ glue-wg mailing list glue-wg@ogf.org https://www.ogf.org/mailman/listinfo/glue-wg
-- It was a sad and disappointing day when I discovered that my Universal Remote Control did not, in fact, control the Universe. (Not even remotely.)
Hi all, I updated the document wrt comments by Andre Merzky. Here: odt: http://redmine.ogf.org/dmsf_files/13260?download= pdf: http://redmine.ogf.org/dmsf_files/13261?download= Regards, Florido On 2014-06-17 11:23, Andre Merzky wrote:
Thanks for the quick reply! Comments/replies inlined below.
Cheers, Andre.
On Tue, Jun 17, 2014 at 11:03 AM, Florido Paganelli <florido.paganelli@hep.lu.se> wrote:
- page 4: 'organization_name' and 'interface_name' are MANDATORY, but '<organization_name>.<interface_name>' is 'RECOMMENDED'. Are there real alternatives to the latter or should that be a 'MUST' ?
See previous comment. It also says MANDATORY for NEW names. Maybe I should stress the NEW?
Ah, I overlooked the 'new'. The statements are correct then IMHO. But yes, emphasizing may make it clearer.
- page 5: 'org.ogf.glue.* SHOULD be used if there are issues with domain names..' - I am not sure what that means. What issues?
for example a domain name is bought by somebody else who is alien to the initial project. Or simply the domain name does not exist.
I just received a request today with domain names that exist but are totally unrelated to the described services.
Listing all the issues would not make sense. Is it a bit clearer now? Do you have better suggestions for rephrasing it?
Maybe instead 'if there are issues' use something like 'if a domain name does not exist or is not applicable'? Either way though, thanks for the clarification, makes sense.
Thanks, Andre.
Regards, Florido
On Mon, Jun 16, 2014 at 4:41 PM, Navarro, John-Paul F. <navarro@mcs.anl.gov> wrote:
GLUE WG:
From our last GLUE WG teleconference June 3, 2014:
Action 2: GLUE WG to review Florido's revised Enumerations document Action 2: At the next GLUE WG call we will vote on the updated document
We will be taking a formal vote tomorrow, June 17, to accept the "OGF GLUE2 Enumerations procedures and best practices" as our practice.
Please review the document here: odt: https://redmine.ogf.org/dmsf_files/13260?download= pdf: https://redmine.ogf.org/dmsf_files/13261?download=
If you have opinion on this best practices document please reply with:
APPROVE - if you support this best practice DISAPPROVE - if you object, please clearly summarize your objections so that we can discuss them
If you approve but have suggested please reply with your approval and list your suggestions.
Regards,
JP _______________________________________________ glue-wg mailing list glue-wg@ogf.org https://www.ogf.org/mailman/listinfo/glue-wg
-- ================================================== Florido Paganelli ARC Middleware Developer - NorduGrid Collaboration System Administrator Lund University Department of Physics Division of Particle Physics BOX118 221 00 Lund Office Location: Fysikum, Hus B, Rum B313 Office Tel: 046-2220272 Email: florido.paganelli@REMOVE_THIShep.lu.se Homepage: http://www.hep.lu.se/staff/paganelli ================================================== _______________________________________________ glue-wg mailing list glue-wg@ogf.org https://www.ogf.org/mailman/listinfo/glue-wg
-- ================================================== Florido Paganelli ARC Middleware Developer - NorduGrid Collaboration System Administrator Lund University Department of Physics Division of Particle Physics BOX118 221 00 Lund Office Location: Fysikum, Hus B, Rum B313 Office Tel: 046-2220272 Email: florido.paganelli@REMOVE_THIShep.lu.se Homepage: http://www.hep.lu.se/staff/paganelli ==================================================
Florido Paganelli [mailto:florido.paganelli@hep.lu.se] said:
I actually don't like RECOMMENDED there, but as a matter of fact we have names with no organization prefix. My choice would be to keep old names, but for new names that have no organization prefix we should force "org.ogf.glue." as an organization prefix to keep consistency. I think Stephen was against it, AFAIR. Stephen?
Fundamentally the prefix is just a convenience to avoid name clashes, so I don't think it should be absolutely required, but there should be a good justification, e.g. that the name is widely known and used. Stephen -- Scanned by iCritical.
Hi all,
Fundamentally the prefix is just a convenience to avoid name clashes, so I don't think it should be absolutely required, but there should be a good justification, e.g. that the name is widely known and used.
I agree with Stephen. Specially for DPM and StoRM, that are widely known and used. It's a bit unnatural to me to call them now org.ogf.glue.dpm or org.ogf.glue.storm. Moreover, they currently publish DPM and StoRM, so it's not a good idea to change it. Regards, Maria
Hi all, On 2014-06-18 10:40, Maria Alandes Pradillo wrote:
Hi all,
Fundamentally the prefix is just a convenience to avoid name clashes, so I don't think it should be absolutely required, but there should be a good justification, e.g. that the name is widely known and used.
I think I am officially giving up trying to have a consistent naming. Is dragging on since two years now. ARC tried to behave and rewrote all the GLUE2 names to be consistent with the naming scheme. It seems to me that most people coding GLUE2 never read the specs -- not even those who WROTE the specs themselves!
I agree with Stephen. Specially for DPM and StoRM, that are widely known and used. It's a bit unnatural to me to call them now org.ogf.glue.dpm or org.ogf.glue.storm. Moreover, they currently publish DPM and StoRM, so it's not a good idea to change it.
Then I suggest we simply use dpm storm At least let it be all lower-case as it is mandated by the schema document (Appendix B in GFD.147 says "enumeration values MUST be lower-case"). SToRM is absolutely not acceptable for me, since the schema is case sensitive (remember my comments on case?) what do you guys think? what do developers think? Cheers, Florido -- ================================================== Florido Paganelli ARC Middleware Developer - NorduGrid Collaboration System Administrator Lund University Department of Physics Division of Particle Physics BOX118 221 00 Lund Office Location: Fysikum, Hus B, Rum B313 Office Tel: 046-2220272 Email: florido.paganelli@REMOVE_THIShep.lu.se Homepage: http://www.hep.lu.se/staff/paganelli ==================================================
Dear Florido,
I agree with Stephen. Specially for DPM and StoRM, that are widely known and used. It's a bit unnatural to me to call them now org.ogf.glue.dpm or org.ogf.glue.storm. Moreover, they currently publish DPM and StoRM, so it's not a good idea to change it.
Then I suggest we simply use
dpm storm
At least let it be all lower-case as it is mandated by the schema document (Appendix B in GFD.147 says "enumeration values MUST be lower-case"). SToRM is absolutely not acceptable for me, since the schema is case sensitive (remember my comments on case?)
what do you guys think? what do developers think?
I guess this will be fine for them. I'll forward this feedback to them. Regards, Maria
participants (5)
-
Andre Merzky -
Florido Paganelli -
Maria Alandes Pradillo -
Navarro, John-Paul F. -
stephen.burke＠stfc.ac.uk