256831 - Address book should be a standalone project

Status: NEW

(MailNews Core :: Address Book, enhancement)

Description

[Domenico Zucchetti]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.1) Gecko/20040707
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.1) Gecko/20040707

Address book should be a standalone project that is maintened and distributed
separately. 
The address book application (very useful) will probably be improved.
The addressbook data should be kept in an XML file format. 


Reproducible: Always
Steps to Reproduce:
1.
2.
3.

Comment 1

[OstGote!]

This is an enhancement, please change the "Severity".

Comment 2

[Wayne Mery (:wsmwk)]

> The address book application (very useful) will probably be improved.
> The addressbook data should be kept in an XML file format. 

It's not clear what you are asking for.  Are you asking for a new data format
(similar to TB bug 266891) or that address book be removed from product=mozilla
to a new product=address book? If the later, what big advantage will be gained?

Comment 3

[Domenico Zucchetti]

I think that it would be interesting to have a complete separate program to
manage address. It would be probably more easy to add other information and
other functionalities. 
The separate program to manage address should preferably store the data in XML
format so that other software (like the mail program) can directly access and
update the file.  The XML file could also easily used to send and exchange
address information.

Comment 4

[Wayne Mery (:wsmwk)]

confirm, to seek developer comment

Comment 5

[Lee_Dailey]

howdy y'all,

i agree that a mozilla address book [or contact manager] would be right handy. then contacts could be handled by one standard program can could be called by the other moz programs like sunbird, thunderbird, firefox, etc ...

as long as the abook program used a standardized interface, then things could be handed back-n-forth as desired without worrying about the exact format of the datastore. THAT would allow extending the info in the datastore without breaking the client apps. 

rather like what apple does ... "one address book to serve them all".

take care,
lee

Comment 7

[Hendrik Saly]

Are there currently plans to create a standalone address book application?

The address book is the most weak part of thunderbird and i think a standalone app  that support a variety range of import/export formats, ldap read/write access and syncml features will be the best solution to build up the address book component to an industry ready application (like sunbird/lightning for calendar purposes).

Comment 8

[Wayne Mery (:wsmwk)]

some recent discussion on the issue can be found here news://news.mozilla.org:119/[e-mail]

Comment 9

[Răzvan Sandu]

Hello,


IMHO, the present Thunderbird AddressBook should be a separate component in the
Mozilla line of products.

AddressBook should exist in two forms:

- a standalone application (similar to Mozilla Sunbird);
- an extension to Thunderbird (similar to Mozilla Lightning), but preincluded
in any installation of Thunderbird (as TalkBack is now).


This type of architecture should offer the following advantages:

- tighter integration with standards for storing and exchanging personal data
(vCard 3.0 - please see http://www.imc.org/pdi/)

- the possibility to establish a FHS-compliant, common place for addresbooks on
the system, which can be *shared by applications* such as Thunderbird, mutt,
Gnome Contacts, etc.

- users on the *same system* should be able to share addressbooks, totally or
partially;

- users should be able to *share addresbooks over the network*, totally or
partially, in a secure manner;

- the existence of a FHS-compliant place for storing addresbooks will allow
further development for software that syncronizes addresbooks with mobile
phones, PDAs, etc. - a facility that is very poorly represented in the free
software world at the present time;

- the possibility to write good-quality converters between various XML schemas
for storing addressbooks (a "bridge" between Netscape-like addressbooks,
Microsoft ones, etc.)

- independent developing (programming) schedules for AddressBook and
Thunderbird;


Regards,
Răzvan

Comment 10

[Răzvan Sandu]

Additionally to my comment above, I will say that we can't expect any mobile phone manufacturer to develop software for syncronizing mobile phones with Mozilla Addressbook as long as we don't have a standardised, public interface to offer to these client programs.

That's why programs like Nokia PC Suite or Motorola Phone Tools, that come with *every* cell phone of these brands only offer PC syncronization with Outlook and Outlook Express...  :-(


Regards,
Răzvan

Comment 11

[Mark Banner (:standard8)]

(In reply to comment #9)
> This type of architecture should offer the following advantages:

All the suggestions (except possibly the last one) that you give for advantages are things that could be achieved without splitting the address book into a separate application. For example sharing address books could be done for users in the existing architecture if someone actually volunteered to sit down and write the code for it.

> - independent developing (programming) schedules for AddressBook and
> Thunderbird;

This is the main thing I don't like about this. You're proposing we have 2 address book deliveries - so we'd have to develop, test, release, QA each of those before delivery with whichever versions of SeaMonkey/Thunderbird. Whereas currently as we are tied in with Thunderbird & SeaMonkey we get twice the QA for free (ish).

Given the relatively few developers for Thunderbird & SeaMonkey, splitting them up even more would do more harm than good. If more people start helping to develop and improve address book (i.e. > 2) then this could be reconsidered, but I don't see that happening in the near future.

(In reply to comment #10)
> Additionally to my comment above, I will say that we can't expect any mobile
> phone manufacturer to develop software for syncronizing mobile phones with
> Mozilla Addressbook as long as we don't have a standardised, public interface
> to offer to these client programs.

Define "standardised". We already have a public interface that both PalmSync and MozPod use. IMHO The main reason those manufacturers haven't written sync programs is that they think there is no demand for it (lack of market share/lack of users asking). If there is a standard interface that would could easily odopt/define there's nothing to stop that being written into the address book whilst it is incorporated into Thunderbird and SeaMonkey.

Comment 12

[Răzvan Sandu]

(In reply to comment #11)

> Define "standardised". 


Well, I'm not a programmer myself (otherwise I would volunteer), but I think that are 2 aspects here:

- software interface for syncronizing mobile devices, independent by manufacturer

- the *schema* of the addresbook itself (fields that don't overlap in Netscape & Microsoft implementations).

IMHO, the actual schemas are "scarce" when it comes to fields - for example, in many implementations there are arbitraily imposed limitations such as:

- one cannot have more than two fax numbers, one in the office and one at home
- one can have a home address and a business address, but not more;
- one cannot attach a tag to a postal address, to say, for example, "business postal address - secondary office in Los Angeles"

The most annoying part is that there is no 1-1 corespondance between Addresbook fields in Netscape implementations and Micro$oft ones.


Răzvan 

Comment 13

[Răzvan Sandu]

IMHO, there is more to add here, such as:

- You've mentioned Palms. Palm devices are *great*, but they are far less common than regular mobile phones (such as the Motorola RAZR family) or smartphones (such as the Nokia E61i, E70 or Motorola Q9). I think we must target especially this customer segment, which is very large.

- Please don't forget about mutt, Pine, elm, Evolution, TheBat!, Eudora and many other e-mail programs, some of them trying to achieve cross-platform functionality. Thunderbird is the most popular and most complete e-mail client, so an address book with a well-defined, public format will greatly improve the possibilities of having the contact data in one place and "see" it form many aplications.
Such an integration would give all these aplications, including Thunderbird, a *great* momentum over Outlook and Outlook Express...

- Above, please don't forget about *text-mode* mail clients, from which mutt is particularily good and representative;

- One *don't have to install Thunderbird itself* in order to have access to contacts data (expecially if an integration as proposed above is attained). *The addressbook should be independent of the e-mail client we use.* Even Microsoft felt that when they created Windows Address Book and they made the same repository accessible from Outlook Express.


- IMHO, the most needed function of the Addressbook application is to import/export from/to a standardised "business card format", probably vCard 3.0.
A format that many mobile phone manufacturers *already support*...



Regards,
Răzvan

Comment 14

[catdogbeloved]

Re: limited number of programmers

This is a construction problem that can be addressed at a later time.

Re: Original proposal to have a stand alone Address Book similar to Apple's own.

I enthousiastically vote in favour of it.
In addition to the list of desired features, please bear in mind the following:
the address book includes personal data; as such, it must meet international
regulations on data protection (European Data Protection Act and the
Safe Harbor agreement with USA and other non European countries). This reduces
essentially to the followings:

1. The stored data must be encrypted
2. Access to the database must be protected by a password
3. Acquisition of data in professional organizations must be authorized by the person, and other details that needs to be sorted out.

The first two requirements are easily met using existing technology in Mozilla.
The third requirement consists essentially in burocracy that professional orgs already know about.

Bear in mind that the above holds for the existing address book too, the one build into Thunderbird.

Bob

Comment 15

[Svein]

My vote is all in favour of this.

At the moment, a common addressbook for all applications is up for grabs. There has been a few attempts at doing this, but none that really works.

I really hurts me to admit this, but at the moment, Outlook has the best addressbook on Windows. Why?

I have about 1500 addresses stored there now. I want ALL of this information - every single field - to be transferred to whatever application that will replace it.

I use Plaxo to keep as many addresses updated as possible. I know that for some people Plaxo is the devil, but it really saves me a lot of manual maintenance work.

I sync my addressbook with my Windows Mobile cellular. This is the most important thing right now.

And on top of this, Thunderbird can not access it so I have to maintain two different addressbooks. Yes, I use Thunderbird for mail. The single, biggest feature of Thunderbird? The ability to set up a list of SMTP servers and designate one of them as default. I move around a lot and have 20+ mailaccounts to check. It would be a nightmare to have to change the SMTP server for each account each time I move to a new network (yes - I know a lot of tricks to go around problems with blocked SMTP servers, but the fact still remains).

So I would LOVE to see a standalone application for addresses. And I would like to see it completely removed from Thunderbird. One less thing to think about for the developers :-) Having one integrated in Thunderbird would just create extra work. Having it completely separate would be the push needed to get things to work properly and we would suddenly have the application we really need.

This would then be a good argument to ask other developers to interface to this addressbook.

One thing I really love in The Bat is that the addressbook also holds templates for messages. You can have templates for mail, reply and forward for each person and each group. But this is just candy. The main thing would be to get full focus on addressbook as a separate application.

Of course, if this does not get the support of the other Mozilla products, we can just forget it. The success of a product like this lies just as much in its adaptation as in its features and development. If Thunderbird suddenly becomes totally dependent on this addressbook, there will be a huge push to get it to work. And that will benefit everyone else too.

Comment 16

[Răzvan Sandu]

<offtopic>

I was ever amazed that, on Linux systems, we don't even have a common *place* to put a shared addresbook (as far as I know, the FHS standard is silent on this).

As I said in comments above, by "shared" I mean:
- common, standardised format shared between applications;
- shared among users on the same system;
- shared among systems over the network (securely, of course);

Best regards,
Răzvan

</offtopic>

Comment 17

[Rolf Gloor]

A good functional address book which includes all the important features (also multiple tags for sorting) and good support for standard protocols and format, would / could become a defacto standard, or at least a widely support standard across the different platforms (Windows, Linux, eventually even Mac OS).

If that can be achieved within TB, fine.