View source

From ASIWG Wiki

You do not have permission to do that, for the following reason:
The action you have requested is limited to users in the group user.
You can view and copy the source of this page:
'''Batch 2 '''

Mr Sarmad agrees with the guidelines and resolutions but he wishes to recommend and to get clear enough that if two separate different languages which are 100% connected to the Guidelines for the same set of languages should use the guidelines with same set of character. (???!) Reword please

Ms Alexa Raad wishes that the guidelines are clear enough that two registries (barring any policy considerations at the registry level) following it – will arrive at the same conclusions regarding implementation for a particular implementation of the same language registries (regardless of ccTLD and gTLD) will come up with the same conclusions. 

Mr Sarmad noted that they allowed, if the guidelines are vague and 2 people has came up with different list…

Mr Mohan noted that it’s ambiguous. 

Mr Sarmad noted that to get a guideline it should be unambiguously design into the characters.

Ms Alexa Raad asked whether guidelines and recommendation are provided, should they come with the same solution?

Mr Sarmad noted that if  it is the same set of language.

Ms Alexa Raad affirmed.

Mr Mohan noted that there could be a benchmark, if guidelines have been passed at level 2 (which is being discussed) and if two independent registries working on “simple” language using the guidelines will result to the same exact….

Mr Siavash seen the logic in respect of two registries which have exactly the same set of characters and in case decided to adopt the same method of solution which may be _______ unconsciously because new method might be discovered by bundles or the solution that was adopted should give the same results.

Mr Mohan noted what Mr Sarmad had said was that guidelines used should unambiguously resolve in the same sets of characters/same language.
 
Mr Siavash thought what Mr Sarmad meant is if he has the same character.

What Mr Mohan heard was if the guideline was used, , two independent registries implementing those guidelines should arrive at the same character set.

Mr Sarmad suggested 2 ways to go either give the characters set or to give the guidelines, but eventually the characters are very clear but if any character set will not be provided (for good reason) then they can go with the guidelines and if somebody follows the guidelines then they should come up to the same characters set.



Mr Behnam made an opinion that if case one language was supported then that is obvious, but if you want to add Arabic to Teran (?> What s Teran?) there is a need to make  some details that support the Arabic too.

Ms Manal says that it might not been obvious if no bundle on variations, the result is constant

Mr Mohan addressed Mr Sarmad to redirect as this was suggested by him.

Mr Sarmad noted  his initial suggestion is to do characters set at level 2 which is sort of sub set of level 1 but it turns out that things may cause problems.

Mr Mohan noted that Mr Ayman supported on that.

Mr. Sarmad says that there was obviously other points, that instead of having the characters set available (sort of rules and guidelines) in case in level 3 and for each Level and Registry (PASHTU, Farsi and Arabic) at level 3 there would be no problem and when you buy multiple languages in level 2, three “Yeh” and if being used into characters set for level 3 for each language should follow the guidelines or rules to create this larger characters set at level 2, its done independently  and it should be redone.

Mr Alirezah noted that any guidelines that has been decided on is only attested in Registry or Registrar of the language in relation to the user because and it cannot influence domain and the GMS protocol and there it’s a must to take care of the guidelines.

Ms Manal disagreed with what Mr Ayman said earlier/yesterday.

Mr Mohan noted in Mr Ayman’s presentation yesterday that he was suggesting to a kind of a top down approach to define language/description wherein a table was build up and as a group decided to go the other way.  Mr Mohan further noted that they must be as generous as possible as well as with the restrictions while going through different levels, perhaps that’s distance of the approached is.

Ms Manal was requested to ask Mr Ayman directly and Ms Manal do so.

Mr Ayman noted to have a clean list so that there are no room for Phishing, there is no room for whatever abuse and then to examine everything else, very cautiously and generate a draft out.

Ms Manal asked in respect of other languages.

Mr Ayman pointed out that they need to make a decision that one language is allowed and the other language is not.

Ms Manal noted that we haven’t solved the 3 languages yet and Mr Ayman replied how to solve the rest otherwise? Simply certain character is not to be allowed.

Mr Behnam noted that in one language in Pakistan they used different character codes.

Mr Ayman reiterated in respect of clean list for the one that should make it to the domain to be reviewed by everyone, there should not have a confusion of dissimilar characters then to generate this other language level list which is very big or whatever is require within the sub domain. He believes that it a duty at level 2 to generate the clean list as possible for those who want to mark it or to register domains that are understandable by everyone across the globe.

Mr Behnam pointed out that such a list cannot be done.

Mr Mohan requested to wrap this particular topic up.

Mr Sarmad noted that if that Yeh, Kaf and Alef will be taken up, he thinks that it can effectively do a lot as there are only 3 vowels and taking 2 of them are being taken out and surely going to cause a lot of…

Mr Ayman noted that there are other 44 languages that are common, than the 4 characters.

Ms Manal supported Mr Sarmad and in case they cannot just convince to register 3 or 4 or whatever characters even if these are not valid and they are frequently used contrary to what Mr Ayman suggested to take away all….

Mr Ayman  replied only to put them in the yellow chart, things that calls for reconsideration.

Mr Mohan pointed out from GPLEs point of view, in such a thing actually happened and that basically going to vanish GPLEs and going to restrict the words that can be registered by users, real words, that in some cases words that have trademarks, associate with names of companies, things like that will be able to register because they are not yet in the clean list and there is no procedure to take them from the yellow list because somebody is always going to have an objection to some points.??

Mr Ayman asked whether there is a means that this solution is going to be handled on the registry level.

Mr Mohan draw the worst case example, its not handle at layer 2 and layer 3, one registry decides and just include all of this and all the finishing issues, that have brought up, happened in one registry because we didn’t do our job here.

That’s what Mr Sarmad wish to point out to adopt that then these group solve the problem and register will come up.

Mr Al Zoman noted not to try to solve the problem now as this problem is complex and not easy. There is a need to work on Layer 2, to identify the characters to be used in the script and identify the problems and try to find the solution, as problem cannot be solved in one day.  Try to find a solution at a script is level, whether it could be reached or not that question is still a question, whether we could find a unified character which is a good solution, there are a lot of things open and haven’t decided yet. It is a complex problem so what is required in level 2 is a task.  There’s a need to work on it and to understand it well to arrive at a decision. There are some characters which are easy to select and these are numbers of character which have some problems, to work with it and find numbers of solutions.  Randomly from the user’s point of view, users may have lot of problems, hosting, costs, configuration cost, once the problem has been defined, define an alternative solution and start proceed from there.

Mr. Mohan noted.

Mr Raed gave his proposal regarding the issue. He noted that whatever done yesterday is an excellent work, the disallowed and allowed characters have been defined in the IDNA Protocol itself and make it extendable to focus on something the norm will use it and define small set of things to be disallowed was done yesterday which is a good step.

2nd – whatever decided on solution which will be provided in case of one Sindhi people, first thing to do is to find the character with the IDNA, to check each language, what are the allowed characters and what are the characters that may confuse for each language itself.  In case of Arabic, an accepted character set and this is confusingly similar, same thing for Persia, Urdu, Pesto and whenever a new language will come up with this community and requested these two tables and at the Registry in case Pesto want to register domain name and they only give a domain name and choose one of the available language a day he can register both, in case he wish to register Sindhi it cannot be registered now and there will be problem in the future, as nobody knows. 

Mr Raed further suggested that bundling should be transferable for the user.  In case someone wishes to register another domain within the same bundle of list, it is another procedure.  A domain can be registered, activate bundle service or whatsoever, based on the allowed/disallowed language that was chosen when registering.  The scenario is extendible, so whenever a new language comes they should develop the 2 acceptable characters and the confusingly character, that’s it from Mr. Raed’s point of view.  From the register, it would be good also to have wide script to read these tables and then apply the rules from each language.

Ms Manal wish to support what Raed has explained as a clean list of a union of all the language/characters. The characters that are confusing is more of a union of all the characters and there a need to check on confusingly character and any new language that is going to be integrated to the table, evolve on the same when using the same character, if the codes are different than one _______, its either have a different form or is confusingly similar.

Mr Mohan noted that in this case level 3 is required.

Ms Manal affirmed.

Mr Siavash suggested doing an exercise to identify or categorize the problem area that needs to be resolved.

Mr Sarmad noted second to the opinion noted by Dr Al Zoman, the need to move on for discussing this topic and apparently there are possibilities that discussion may not close for another extra period of time and suggested to make a list of all the issues which relevant to this level and see the possible guidelines/recommendations/solutions which can be defined and once sub listed it can be reviewed and close. What are the potential issues can come up and what potential issues should have.

Mr Ayman noted that they are now talking on the same language, though they are focused on different areas:

Mr Siavash wishes to define the issues;
Mr Al Zoman wish identify the tasks; and
Mr Sarmad wish to find a solution or how to solve the problem.

This is what he meant in yellow list, (clean list, no problem, solved) wherever there is certain addition which answers the low concern for the moment. Yellow list is not just to take decision but to pass to level 3 to work on solutions and this might take time and but if a  solution will be defined it would be part of a guideline.

Ms Manal noted that there’s a need to identify the issues, problems, regardless the means of time from the very high prospective confusingly similar characters/languages is an issue, combine character is an issue, from the very high prospective.  The main issue is that this might be applied to other languages.

Mr Mohan wish to proceed on the issues to get on with knowledge and do that and restate what he has heard so far.

A proposal and a model that is workable that says, for each language to be included the language clearly or provide list of acceptable characters and confusingly similar (variant) tables. Once they proposed and provide that then this group __________ when comes together looks would needs to happen and then comes up with recommendation at that maybe a general purpose guideline should look.  He is identifying this proposal as a candidate guideline. 

The pure request is to start talking about one of the major problem areas that needs to handle and then to start actually discussing it.

Mr Mohan listed out the following topics to discuss and as suggested by other participants:

===•	Confusing similar:===

1.	Kaf Group<br>
2.	Yeh Group<br>
3.	Heh Group (1)<br>
4.	Heh Group (2)<br>
5.	Alif, Hamza and Mahda<br>
6.	Numericals<br>

Mr Raed would like to propose in respect of Arabic, these are the acceptable characters, if someone register the domain name all these will be blocked or whatever (referred to the Chart).  From Arabic, in Kaf there’s a lot of confusion in respect of Kaf, \number and so on.  There are some general issues; if it can be defined it can determine whether it is for the Arabic/Persian, etc.  A table can be put, however for the extendibility purposes this can be joined because there’s a need to remove language and these two should be provided.

Mr Sarmad noted that that is one of the solutions.

Mr Mohan continued listing out:

===•	Combining Marks===

1.	Tashkeel<br>
2.	ZWNJ/SWJ<br>
3.	BIDI Issue<br>
4.	Consisting Multiple Words<br>
5.	Delimeters<br>

Mr Ayman noted that these are as far as the 3 languages are concern not the 50 languages.

Mr Al Zoman suggested starting per language.

Mr Ayman pointed out the need to define the proper solution, how to match the solution with the issue.

Mr Sarmad suggested to take one a time and identify what the problem is and then document and possible solutions. 

10:30 – Break

Resumed:

Mr Mohan noted that there was a proposal that composed during the break which is to actually review some of the proposed solution or at the list of the proposed solution before going into the reviewing the issues that are listed.

Mr Benham wish to propose a solution which they have experimented somewhere else not in IDM and wanted to make a presentation of the same.

Mr Mohan asked everyone if they are agreeable to  Mr Benham presentation.

Ms Raad agreed.

Mr Siavash asked whether it s a general solution to the entire problem?

Mr Benham noted that it is solution for some cases.

Mr Sarmad pointed out to Mr Benham to make a brief focus on the problem/issues.

Mr Mohan noted and given a 5-minute presentation.

Mr Benham proceeded with his presentation:

Mr Benham started and it is about a bundling system to be use for some of these problems.  There’s a need to know how many number of variants? should be for each domain (Mr Benham draw an example) he noted a requirement to know how many fvariants each bundle are going to generate, the number of variantz defends on the domain itself (maximum).

Mr Aliresah stated that if one of the members of the bundle exceeded as the domain ______ needs to drop it off.

Mr Benham wish to state and to be part of the recommendation that in allowing to register a domain that some of their variantsshould pass the domain.

Another issue is about Harakat (Mr Behdad started with the presentation in respect of loose searching).

Mr Benham noted that in this case he wish to suggest all the registries should talk to the registrant first if they can register to prevent fishing, it is up to the user to register or not.

Mr Raed noted that this is not within the scope of the meeting and wanted to focus in the exact similarity not something that the user may confuse.

Mr Benham replied that what he is trying to inform are about language problem and wish to note that  (i.e.,) in registering Persian domain it should focus to the user that he can prevent fishing while registering the other one.

Mr Al Zoman noted that this is part of the discussion which was already been done even in the Arabic language, connector folding between Hah, Wow with Hamza, Alif and so on.  It has been decided at the level of supporting the language is not folding and if folding will be done just for the purpose of like typo mistakes or laziness of users, these could be solved at the registry.  This problem in respect of languages, this is in the high level this is not at Registry level, for him it should not be even part of the guidelines, this could be added value from the Registry, in trying to solve the laziness of people. 

(Mr Bendad continued the presentation in respect of Harakat)

Mr Benham noted that the solution is to disallow registering Harakat. To register one with Harakat and one without Harakat.

Mr Mohan asked Mr. Benham whether what he is proposing is that these are the things that should not make it to level 3 and this stuff should be done at level 2.

Mr Sarmad understand that Harakat is a significance issue and maybe not telling things based on the assumption so far that they are required by some languages and may confuse the standard or IDN in the process.

Mr Michael noted to advise people that though they are part of autography they are not allowed in IDN.

Mr Sarmad explained that assuming that they are part of autography, we don’t know that yet.

Mr Michael noted to Mr Sarmad that even in Urdu have some obligatory use of…?

Mr Sarmad affirmed that could be discussed and for the larger _______ there are something which could do to compromise that has something to go back and discuss more detail but just saying that maybe required by one particular language so the whole IDNs system for Arabic Script has to accommodate that or should be opened to both possibilities.

Mr Benham is suggesting something that doesn’t make any problem for languages that don’t use Harakat and if you wish to register this no one can get any of this so there would be no problem.

Ms Manal asked in respect of Kasmir whether it is within the language or they just use it?

Mr Aliresah thinks that aside from bundling process there’s also a need for normalization process and noted that what Mr Benham is trying to emphasize is kind of normalization process  which need to think about align with bundling.

Ms Alexa tried to understand what Mr Benham is trying to propose is in relation to level 2.  She noted the issue in respect of trademark, either in normalization or bundling, we have Malik and Mulk. It can actually be written at the same way but in removing that Harakat (Malk) is a generic name and Malik could be the trademark.  In this case it is not a problem has been created in trying to solve one problem. Ms Alexa asked if something has been created. And if so, if trademark issues are part of this and talking about opening up an IM Registration, there a need to think about trademark and to how to solve it both.

Mr Benham replied that it depends on the language.

Mr David says that UDRP will take care of it.

Mr Alexa raised an example when something has been blocked either on registry level or level 2, that happens to be trademark that the group did not know about, and noted that UDRP will take care of it at the Registry, you’re not held liable?

Mr David – I can’t believe that we would want to get into that ________, we probably do if we created system and prevent someone from using a trademark and domain name.  I don’t think it is sensible for us to try accommodating trademark.

Mr Al Zoman noted domain name are not trademark.  

He proceeded to Tashkeel (Hakarat) and pointed out that it could be allowed in the application level, the names can be pronounced good but it is not as a distinguished character, whatever Harakat is required and only the basic character will be taken.

Mr Benham asked based on what Mr. Al Zoman said that he is suggesting to disallowing all the transferable character in all level.

Mr Al Zoman replied only in Harakat and he is talking about the points, the definition of Unicode.

Mr Benham noted that languages using Harakat are mandatory and not forcing everybody to support Harakat in ccTLD or some GTLD, but if GTLD wants to support _________, it should support Harakat, bundles for these two.

Mr Al Zoman noted that they are not talking about the language but the script, for the users who are using the script as to how to protect their rights and make it easy for them.  With Harakat the cost will increase at the user level, registry and hosting level, along with the confusion.

Mr Benham asked no bundling?

Mr Al Zoman replied that bundling can be done.

Mr Mohan asked Benham whether he wish to proceed with the presentation or reply to the respective objection in queue.

Ms Manal noted most of the points were covered by Mr. Al Zoman, it also depends on the script and asked the need of Harakat in a day to day writing, in registering a domain name with Harakat it can access to it without Harakat.

Mr Raed noted that Tashkeel/Harakat was disallowed from the Protocol level. Harakat is a problem and all of us do not support Harakat. He further noted that let say at the registry level we say that Harakat can be supported in higher level, but in the registry itself there is no Harakat, in the future we need harakat, then we can apply policies or procedure to reduce the fishing, but now we already agreed that Harakat is not require for each language and it can be handled by higher level.



'''Batch 3'''

3rd Issue – is that even if they need Harakat it can be supported by higher level, in the registry they can register without Harakat and then they can have whatever Harakat they want because they fall with the same domain name.

Last issue – using bundling with Harakat will give an unlimited number of combinations, it will be costly for the user, registry and others.

Mr Benham noted that there would be no problem when Harakat will not be used? But in case that it will be named under the Kurdish and Teran, Harakat will be needed.  So there’s a need to decide what to do in level 2 not in level 3.

Ms Raad noted to Benham that in case Kurdish need to be supported acceptable Harakat can be shown to them then they can put Harakat and noted that this might be confusing with the Arabic. 

Mr Benham pointed out that Harakat cannot also be allowed in TLD and users can register without Harakat at all and there would be no problem.

Mr Michael noted that Kurdish doesn’t need Harakat, they tried long U, and they write Wow twice. There is no guarantee that domain name is correct and suggested that this should be kept close until such time somebody comes.  They might need to learn that in domain name that these things are not written.

Mr Sarmad noted that since they are already allowed at the protocol level, and if it will be allowed (Harakat) now and it can never go back.  If this will be disallowed them it can always be added later with more mature policy in the future, when the needs arises, we can amend guidelines and include.

Ms Manal affirmed either mature policy or review periodically.

Mr Mohan seen lot of heads nodding in respect of the proposal that in the first version of the guidelines that can be recommended that Harakat not be used in the acceptable character set, either pressing need or regular review will amend this guideline and to keep it now and keep it open for later.

Mr Al Zoman agreed that this is what they required, let it out from one level and made the solution.  From the time being this will not be supported but later on if there’s a need it can be amended.

Mr Mohan noted that this is the 2nd guideline candidate.





Ms Manal says that guidelines should tackle everything, should bring to the registry’s attention everything to be taken care of.

Mr Mohan noted that that is an opposing view.

Ms Tina Dam says that it can be used at the registry level, but cannot be used at the protocol level. It is a linguistic position that doesn’t sit at the protocol level its something that sits on linguistic…

Mr Sarmad noted the requirement to have it defined at the guideline level as noted by Ms Manal because there are many things ________ which has not implemented and on the registries which will only be implemented. There is no way back but if we restrict it now, one year later or whatever and the needs arise this can be addressed. This is a more flexible solution.

Ms Tina Dam suggested not disallowing and instead put it in pending category.

Mr Al Zoman noted that in the Initial Presentation of IDN using the Arabic Script this group of character (Tashkeel) should not be supported and if it needs arises, there should be a justification.

Mr Mohan noted Manal’s view, the guideline will be reviewed periodically or if there is a pressing need.

Mr Michael says that when the pressing needs come it will be explained, but asked whether considering the consequences.

Mr Al Zoman pointed out if they could come up with a good justification that that is needed.

Ms Alexa Raad noted the consensus on disallowing them for now unless there is a pressing need and further noted wasting a lot of time on this topic.

It cannot be solved here as per Ms Tina Dam and its better to put this at the pending category as she mentioned earlier.

Mr Mohan actually heard with the exception of what Mr Siavash is that Harakat be not supported at level 2?

Mr Siavash noted that it all started when Benham made a presentation.  He argued this is going to stop possible technological opportunity as to how to cope with this problem as he don’t see any problem at this time and off clue whether it should ban something that do not know the consequences are.





Mr Mohan heard what she’s saying is supporting Mr. ___________ was saying as well as she tried to be as restricted as she need to be.

Ms Tina Dam  said no because of level two they have changed, but level 1 they cannot.

Mr Sarmad noted that banning is not the right word, he don’t think that is intending to be banned.

Mr Mohan noted that proposal says that Harakat should not be supported at Level 2 to be reviewed in guidelines or pressing needs.

Mr Sarmad suggested that they may want to soften that to get everybody on board.  In Phase I Harakat will not be used, in Phase II if there is a pressing needs, it maybe considered or continue with what they are ought to do.

As per Mr Ayman they can also add the existing language groups.

Mr Mohan pointed out to everyone that if they wanted consensus they have to convince Siavash.

Mr Siavash says that he is not trying to be hard headed but it just seems to be a clear adopting of something ……..

Ms Manal pointed out that when they came to a conclusion or to a recommendation this is more of trying to share their best practice with other registries who might not be aware of or who didn’t have a change to try and its not banning as Sarmad said.

Mr Mohan informed that the proposal that are currently on the table is that, the existing language group proposed that Harakat will not be allowed in IDN in Phase I of implementation of IDN, this recommendation will be reviewed in future guidelines or when pressing needs demonstrated.

As per Mr Siavash he would not propose such a thing.

Ms Alexa Raad thought that the word ban is very final, permanent, allowing something which is very firm that may end to a wrong decision to pull it back.  Considering that there is no enough data and it is better allow it for now until there is a demonstrable phase, potential issues, then it will be opened then support.



Mr Siavash thinking of solving this problem in a way that is more comprehensible or satisfactory that could be reached.

Mr Al Zoman noted that it is not a matter of stopping them, in Phase they can go on what they can, Phase II if this is really needed it will be supported and make a solution and address it, but not avoiding the problem.

Mr Baher raised a question whether if the feature is not supported in Level 2, does this mean that it (will be or will not be) supported in the higher level?

Mr Mohan replied that that would be the implication.

Mr Baher raised another question, this current guideline whether they have the solution in respect of Tashkeel, IDN in Level 2, had this been addressed? He noted that maybe they are trying to prevent something that is already there.

Mr Mohan informed that there is no enough data for confirmation and that is a general issue.

Mr Sarmad believe that he is getting a little convinced with Siavash, he thinks that there is need to make their wording more literal, it should actually say that we are not allowing Harakat not because they are bad but because the solution is not yet known and will find solution and then will consider whether the Harakat should be in or not.

Mr Mohan has a strong objection of what Mr Sarmad is saying and what Mr Sarmad is proposing is actually not a guideline, its got to be are real guideline.  Mr Mohan noted that they are proposing not to support it but as soon as they come up with a better idea then it will supported.  Further pointed out that it is their responsibility to be asked clearly.

As per Mr Sarmad that is basically he is saying, don’t include them now.

Mr Mohan wish to unmark his objection, will add the extra explanation but I still think the need to do something solid that can take away as guideline or rule… existing language group proposed that Harakat not be supported at Level 2 in Phase 1 of IDN deployment.  That is remained to be strong.  The second sentence would be, to be reviewed and guideline when pressing need demonstrated.

Mr Sarmad noted that they are no longer taking about pressing needs, there is a need to go back, collecting data, find out and propose for a solution.





This would be a modification of the proposal as per Mr Mohan.

Mr Siavash/Sarmad commented that other characters become struggle. 

Mr Ayman says that instead of the pressing needs incident, an urgent solution.

Mr Sarmad affirmed.

Mr Sarmad noted that economically speaking, they should restrict the words and  considering all options.
Ms Manal wish to point out that what Mr Siavash is saying is that they are banning, this going to be reviewed and on the contrary it encourages them to keep on thinking to come up with a better solutions.

Mr Siavash maintained that he haven’t seen any pressing need and why bother to register towards something to which of no instance, there’s no problem yet.

Ms Manal repeated as suggested by Mr Sarmad that they all agreed that they don’t need the Tashkeel, but they don’t have a solution right now.

Mr Siavash noted that it would have been nicer if they could have Mulk, Malik, all of this registered without any confusion and if they would register something that discourages people from finding solution…then they don’t want to find solution.

Mr Ayman replied that it is not that they don’t want, its just that they don’t find a solution yet that they are confident about.

Ms Manal asked it this will be used in a day to day typing.

Mr Sarmad noted that it is empowering the user, user should be given maximum flexibility, whether they use or not it is there because its part of the language. Further noted that if they could get it without confusion it would be nicer.

Mr Al Zoman suggested to start without them and if they have mechanism to allow the user in a usable way then there would no problem.  From the users’ prospective it cannot be done from the time being, as they afford to support them but it can be revisited and so on.

Mr Mohan wishes to stop the discussion of this topic at this point as they may not have consensus and certainly a strong support by this. This won’t go to consensus yet and this may even discuss this on the main list and proceed with the listed topics. 


Mr Mohan asked Mr Benham whether his purpose of the 5 minute presentation is to inform the group or he wish to direct and guide the group in some direction.

Mr Benhan said that it is only information of everyone.

Mr Mohan asked whether has that been addressed.

Mr Benham affirmed.

===Topic to be discussed:  ===

Tashkeel – done.

Combining Marks:

Mr. Al Zoman started the discussion on the above and explained that a combining character is like when you take a basic character and you wish to adopt other characters like Mahda, Hamza and other shapes along with the characters, it will create different sounds and other languages.  Eliminating them is a not a good choice, but we need to understand the side effect of having them in the table.  Some of the characters who have the Hamza combined form and put it over “Alif” it has to be folded automatically with the exact connector.

Mr Michael noted the ones which are chronologically equivalent are.

Mr Al Zoman asked Mr Michael as to what other combination is that group that needs to be addressed.

Mr Michael informed that he can guarantee that 0627 (Alif) + 0653 (Mahda) are identical to 0622.

(A lady arrived - late comer).

Mr Al Zoman noted that this Mahda if you can combine it in basic Alif which is here, it will give this letter (He referred to the chart). Also, Mr. Al Zoman pointed out wherein they have a problem (big dot).

Mr Michael approached the board and drew and says that in African language if you have “Beh” (referred to a dot) its supposed to be bigger.

Mr Al Zoman informed there are many combination if you taking this 0658 with dot like a shape of “Tamania” and put on top of “Wow” it can give other character use in other languages (referred to the chart), he informed that he don’t know if that is normalize or not.




Mr Michael informed that the things that are normalized are things that are already normalized and nothing else will be normalized.

Mr Al Zoman pointed out that some of the characters which is in the table using some of the basic character in the Arabic and add this combining character to them that would cause a lot of problem for other languages.

Mr Sarmad suggested one possible solution in relation to 65C which is a dot, guideline at application layer they reserved that this have to be submitted and they have special guidelines so it will not be confusing with other and noted that is an application issue.

Ms Alexa Raad asked whether they have a recommendation for the application providers.

Mr Sarmad replied Level 4.

Ms Alexa Raad asked it’s neither that you recommend for the Registry?

Mr Sarmad replied that at this point it requires for the language.

Mr Al Zoman suggested to gather all the combining letters and try to add it to some other characters, produce a table and try to find the similar character in the table to see the problem and normalize, to check whatever is ______________ or not and then find a solution.

Ms Alexa Raad asked as to how they would know whether it is handled in the normalization.

Mr Sarmad replied that for GTLD it is a problem, for ccTLD it is not a problem because in Pakistan they don’t use it, for the African Countries which is using it they only use so that they would know of the potential characters that they know from the context whether it is a dot or a “Wow”. The only problem is that GTLD where they are giving larger context in supporting all languages and at that point as a GTLD there’s a need to have a policy that are going to support this language or not.

Mr Al Zoman reiterated to produce a table to determine from it that there is one similar and to suggest which one to be used.






Mr Alexa Raad noted that they are looking at this as the category of combining characters, it sounds like it depends on what combining character they are talking about and whether it could be combined where and outcome will be different whether it is an application or potentially a recommendation to the registry.

Mr Al Zoman affirmed.

Ms Alexa Raad raised a question whether there is a requirement for the combining character?

Mr Bendad replied that Mahda and Hamza should be considered different at least.

Ms Alexa Raad asked whether they have them separate already.

Mr Bendad answered that they can.

Mr Siavash suggested making a list of all the possibilities that might come up and have them available, to determine what can be allowed and if not take a possible measure to address.

Ms Alexa Raad wishes to confirm whether it is a specific combination.

Mr Siavash replied that they can make a list.

Mr Al Zoman – (He approached the chart and explained how)

Ms Alexa Raad asked that when you take 2 characters, combined have a table and figure out from the combined characters, what would happen during normalization.

Mr Sarmad noted that actually the first part is automatically taken care by the protocol and listed out categories below also suggested by Mr. Al Zoman:

1.	Normalisation;
2.	Which produce a connector valid for both but is not covered by normalization; and
3.	Which produce new character that maybe is not use?

Mr Michael would like to avoid using combining marks.



Ms Alexa Raad noted that it seems that combining characters are more of issue of confusion that one of these, because all the characters are look like it (referred to the chart)

Mr Sarmad – there maybe some languages which may be combine with Mahda and some other.

Mr Raed (referring to the chart) as what Mr Sarmad said these two are not use so it can be blocked, we are not using them on Arabic, Pesto or Urdu, etc. He recommend the same thing in respect of Tashkeel.  For instance, in Arabic he will not support this because they have risk for us for (Arab) same thing with Pesto, Urdu, etc.  They should continue with this one (chart) but not support at this stage.  Whenever a language needs a combining character they should come and will provide the table, update the table if they really need it, if they don’t need it they can continue without solution, there would be no problem for that.

Mr Al Zoman explained that when they said table, it is only the script to these three tables, not a language. He never support it anyway at the language level.

Mr Siavash says that they are not using it either.

Mr Raed replied that why they are now focusing on languages that they don’t know and cannot predict whether they are going to read it or not.

Mr Al Zoman suggested bringing the language that share their script, sit with them and then they altogether solve it.

Mr Siavash replied it’s their problem let them do it.

Ms Manal noted to adhere in respect of the kind of approach, they should start with the minimum or maximum.

Mr Sarmad affirmed and wishes to comment on that because as per yesterday they are talking from a different perspective and which they are actually going in a different perspective. What he had in mind was a very literal set of the case, rigid, the protocol, something smaller but still eligible and by way of doing things, they don’t want to overwhelm and everything that something that they have allowed in Level 2 which they think should not be allowed but protocol is allowing it.  But he thinks that what they have done is to all those ______ at protocol level as well, level 2 is probably becoming redundant and what they should have is just a guideline in level 2 and table at level 1 and level 3.  It should be guidelines at 2 and 4 and table at 1 and 3. 



Mr Mohan requested to restate what he said because it doesn’t seem to be cleared and asked him whether what he is saying is that our approach seems to be changed compared to what we discussed yesterday and one of the proposals that coming forward is that we don’t look at characters and restriction at level 2 but instead at level 2 just focus on guidelines and we allowed restrictions to happen in level 1 and or at level 2?

Mr Sarmad affirmed and explained that this is a little disturbing for him because it is going to be a long term commitment, expanding things and then on top of that it brings only different voice of low standard which GPLE would have achieved or ccTLD would have achieved incorporating and improving overtime and this is a very dynamic kind of approach which we arrived at and what I had in mind was a very static kind of approach.
Return to Asiwg 1st meeting.
Retrieved from "http://asiwg.org/wiki/Asiwg_1st_meeting"
View source

From ASIWG Wiki

Views

Personal tools

Navigation

Search

Toolbox
