MailNews Talk:Future of Mail

From MozillaWiki
Jump to: navigation, search

Encryption As A Key Benefit (pun intended)

Hi -

My first ever contribution to a wiki of this sort but it's due to how highly I rate and value Thunderbird.

Thunderbird has made the use of OpenPGP (via enigmail) easy, but it could be and should be even easier. Sending and receiving encrypted email (even that word "encrypted" is enough to put most people off) should be as easy as sending a regular mail. A couple of suggestions therefore:

1. In the address book there should be 2 types of entry - a plain email entry and a secure entry ie an email for which you have the public key. The default would be to send encxrypted mail to all entries for which you have a public key.

2. The Thunderbird address book should in essence do away with the Key Management tool under OpenPGP menu. The user would handle all encryption settings from within the friendly looking Thunderbird address book interface.

3. Thunderbird should automatically include Enigmail/OpenPGP as part of a seamless install. Key creation and key backup should occur automatically.

4. Public keys have to exist, but masking them as a small image, an icon, an avatar, would make the adoption of encryption more playful and user friendly. Users just send an avatar saying "here's my security key". Stock avatars could be provided but an option to supply ones own should exist.

5. Account option so that any mail received that is not encrypted is automatically binned and/or deleted on the server if possible.


Thunderbird has done more than any other client to make encryption easy. I still think it is an area that has massive potential not only in terms of security but also in obliterating spam. I recommend Tbird to Linux, Mac and Windows users alike and assist them with Enigmail and OpenPGP set up.

Paul at thefreepress dot org dot uk

  • TB has extensive support for S/MIME
  • There are several CAs providing free S/MIME certificates.
  • PGP is similar to self-signed SSL certificates (no verification) and Mozilla is making efforts to limit its use. I guess this request is into opposite direction of that effort.
  • Users which have the knowledge about its use (and its limitations) are able to install enigmail as an add-on. This shouldn't be provided for the casual user without the required knowledge.

Eddyn 04:30, 4 August 2007 (PDT)

integrate serverless Email and Instant Messenging

Hello Mitchel,

thanks for this initiative, I think I have two good ideas to the development, the Thunderbird development team and the organizational structure.

Maybe Thunderbird has not such a usersbase and development features and money resources to stay in that shape as it is.

But I would not through away the baby with the bath tub, as we say here.

First, making a new organization/foundation is too much work and takes time. So I suggest to give Thunderbird just one year more in the Mozilla organization and then make it either a child company or a sf.net community project.

After that year, do not miss to ask the developers and coders about their interest, in which environment they want to work.

Please, let Thunderbird for one year in this organizational structure as it is. And we can develop the following ideas.

1. Thunderbird is a very good mail client. It is in one step mass-ready: But, Outlook is still better, because it has a calendar. So just add to Thunderbird a calendar function to get reminders for emails to recipients with a birthday. So instead of outsourcing Thunderbird, merge Thunderbird, with Sunbird. Both are birds, you know? http://sourceforge.net/projects/portablesbird/

2. And this is the new idea of email: 80 % of email we do to trusted friends. And we do less email, because we are on Instant Messengers. So we need like in Google Mail the option for both: sending an email message or - if online - an Instant Message. There is a new serverless Instant Messenger out. this is this project

http://sourceforge.net/forum/forum.php?forum_id=618174

It is open for client and protocol, for all posix available, at the moment with QT/FLTK gui.

The main principle is, that all communication is done ONLY to friends, which were defined before with a symmetric key exchange. This means, communication is safe and encrypted.

This has the side effect, that no one can message and email me - besides the defined trusted friends.

Retroshare Messenger is both: an email client and an serverless Instant Messenger.

So... sending serverless Email is definitely a feature, how mail is in future organized. The same for Instant Messaging. The Instant Messenger market is one with very high interest, see Google Talk or the merge of AOL-ICQ and MSN-Yahoo and all the Multimessengers.

The idea is now: to implement RetroShare serverless Email and serverless Instant Messenger into Thunderbird.

This should be scheduled for one year of work within the Mozilla Foundation (just a gui integration) and then we release some betas of Thunderbird with Retroshare Instant Messenger and serverless Mail and ideally with Sunbird Calendar. The official Release is then a full integration of RetroShare serverless protocol for serverless Email and Instant Message and the Sunbird.


3. Thunderbird is used not so much, because it is not bundled into the Firefox Installer. One Comment said, that the Mozilla Foundation should then use the name Firefox Foundation. But instead of your initiative, I would recommend to a) give Thunderbird one year merging with Retroshare and b) going a little bit back to the roots: Mozilla Foundation was Netscape? So Mail and Surf. And if the we get Thunderbird with Firefox bundled to ONE Suite, then we would not have these problems. Really, I am really surprised about the suggestion to outsource Thunderbird, if it was Sunbird, ok, but Thunderbird? Even Netscape said, that the actual version is only browser. But a few days later they said an email client will be not excluded for the future!!!

My suggestion is:

1. Integrate Sunbird in Thunderbird 2. Integrate Retroshare in Thunderbird 3. Bundle it to the Firefox Installer.

Then after one year of development, decide new and see, how the product is used or not.

And last: If there is a new organization needed for Thunderbird in one year, this is definitely the OpenOffice.org community, as they play around with a calendar and as well with an Instant Messenger.


  • IF* Thunderbird MUST be outsourced now, then make a new joint-venture-company with OpenOffice, for a Product, which has Email (Thunderbird) Serverless Email and IM (Retroshare) and third a calendar (Sunbird) integrated.

But it is a mistake to do it isolated outsourced (merged in a joint venture or not), this is why I suggest to have a very HIGH DEVELOPMENT for the next six months, to get the first version of Thunderbird 2.7.9.9 launched this year, with Sunbird and Retroshare implemented in the gui. The beta should be this year released.

Then in 2008 we get first official Version of 2.8.0.0


I think this would be a good vision to email: a serverless email and message communication done with the retroshare protocol in the Thunderbird.


Besided it would have good synergies to have the option to mail to friend over Retroshare (no spam, confidential mail, sending of documents... etc) and second you can use for Mails to public Mail addresses (with @ in a mail address) the Thunderbird. Thunderbird would get an Instant Messenger as well.

Last idea: Retroshare is discussion to add the www.sim-im.org Multimessenger as a patch to Retroshare, or other way round, SIM will integrate Retroshare.... this both added to Thunderbird would be like a Multimessenger we need. - serbased Mail Message (Thunderbird) - serverless Mail Message (Retroshare) - Serveless Instant Message (Retroshare) - Serverbased Multimessenger (SIM-im.org with AOl, MSN YAHO ICQ JABBER) - Calender Function like in Outlook (Sunbird).

This should be the plans for Thunderbird in the next months.. a lot of stuff to do, maybe you can discuss it with the small development team and get a few coders as well from Firefox.

Remember: The main goal is to get an Installer ready, which is Bundling Firefox with the Message Tool.

Reading Information is only the half of the medaillon, the other half is to discuss them online with friends.

Incoming Information (Reading web, Firefox) and outgoing information (Email, Thunderbird) are 2 ways of communications, which are essential for human beings.

So please do not make mistakes! Give Thunderbird a Push with a serverless Instant Messenger.

Thanks!


make webmail not suck

I'm aware that Thunderbird and Firefox have a lot of code in common. AFAICT, Thunderbird is 90% JavaScript/XUL/CSS etc--stuff which could be served over HTTP such that from your web browser you could almost have a Thunderbird UI as a webmail app. And it's up to the server admin to keep it up-to-date, configure it for his mail service etc..

The (XPCOM, I think) bits that are missing could possibly be offered to the Firefox client as an XPI file, or maybe if they're small enough (given that, I think, none of it is UI so no i18n concerns etc) bundled into Firefox itself.

Or maybe something new needs to be invented for this crazy idea to work. I don't see an obvious way for things like drag-n-drop attachments to work. But I thought I'd mention it as I'm old enough (unlike my student users) to think Thunderbird is far, far better than SquirrelMail.

--Mb 14:39, 25 July 2007 (PDT)

Using Thunderbird/Lightning as a UI for a webmail application

Mb, you should check this out: http://groups.google.com/group/mozilla.dev.apps.calendar/browse_thread/thread/e398d135e7e4a459/92f2ae4540a67b5e#92f2ae4540a67b5e

Next generation thunderbird

Hi there,

Some ideas for the next generation of email. The general tagline for the strategy is "Easy to use information & communication hub".

My idea is simple: we all check email every so often, so why not group everything that needs checking into the application,and add value to email messages via automatic classification, tagging, linking and so on. Let mail messages be more than text in folders!

It's based on four general directions:

1) A much more usable/intuitive GUI. Think Apple.

2) Making it more like MS Outlook: Adding calendering, To Do lists and making Thunderbird a hub for information management

3) Improving searching, tagging, cross-referencing...etc. Revisit the "intertwingle" principle from jwz.

4) Strong integration with everything else: Firefox, The Windows/Linux Desktop. Web sites & other webmail clients. If it adds value it could be either integrated or have a plug-in.

I'm currently sketching some of my ideas regarding a proposed GUI and detailed features in each of those directions. I'll try to post mock up screenshots in the coming days.

update: here's a screen shot I whipped up using Krita for a proposed "Today" screen like Outlook.Sorry if it looks a bit rushed :)

http://samy2004.googlepages.com/thunderbird2008

~~Mohamed Samy

An unfortunate choice of words

Quote: "Mozilla has a range of resources – funds, code, etc.—that can be applied to this problem. We’re looking for people with expertise, vision and leadership capabilities." Geez Mitchel,

That seems to imply a complete lack of confidence in the existing team.

Maybe some of that "range of resources" could have been applied to hiring an additional programmer, possibly in lieu of the 100K grant to ........player

I think you owe Scott an apology on this mis-speak.

Might account for his favoring an independent company for further tb dev I think I would go that way in his shoes.


Mitchell Baker notes: I've had many conversations with Scott. He is quite clear his focus is Thunderbird, the product we know today, and caring for those users. He's committed and passionate about that, and that is extremely valuable.

If we want to see a different kind of mail initiative, it will require someone with a passion for that initiative.

More thoughts

I think that this tends to be too much throwing out the baby with the bathwater, as stated above. I think that Thunderbird can be a great starting point for innovation in mail, but it has to stop following in Firefox's shadow. So, I'd support any solution that helps break this lock-step development where Thunderbird creeps along piecemeal in the shadow of the larger application. I don't know that changing organizational structure will really achieve this. I think it requires a different mindset.

One of the primary things that Thunderbird has always lacked is a calendar. That's one of the things the Lightning project brings to it. Since Outlook became a player on the scene, people have been conditioned to managing a calendars in the context of their e-mail. So, I think that the Thunderbird team and the calendar team need to continue working together to make a kick-ass all-in-one manager of internet information (which in my mind is the number one selling feature of Thunderbird). That is also how I see Thunderbird fulfilling its portion of the Mozilla manifesto.

Now, all this stated, I would be willing to aid the new mail initiative by helping them integrate with the existing calendar functionality. We have a ton of good calendaring code in the mozilla tree, with a decent architecture, and I want this mail initiative to make use of that code so that we can continue to interoperate across all the mozilla products both the existing ones and the ones to come.

The new mail initiative

Is already underway at labs.mozilla.com! The Coop!

I really hope that the coop gets a lot more attention than it has so far. In my opinion The Coop is the perfect solution to innovation in email.

And robots.

Key Thoughts

  • Use mozStorage for storing mail instead of the text-based mailbox format, as well as for storing contacts instead of the old mork format. This should make adding new features easier.
  • Manage all forms of electronic communication, not just email and newsgroups. That includes IM...
    • It would certainly be too much to integrate in a whole IM client (nor should we force the user to use any specific one), but it would be doable to, say, detect my GAIM/Pidgin install and let me search through my IM conversations with a contact just as easily as my emails to/from him. (It's worth pointing out that one developer was able to build a basic AIM client as a FF extension in only a few kilobytes!)
  • ...as well as online forum and blog discussions.
    • This would require some integration with FF, but it would be great to be able search through and read all the past online discussions you've been a part of.
  • Personal Life Recording. With Lightning, we should be able to record all sorts of time-related data. Something like Qlockwork would really help keep track of what I've been working on and remember details of my life that I'd otherwise soon forget.

Some of these might be best left as extensions, but to me at least they'd be killer features.

Note to others: there already is an app that integrates mail with the browser: Seamonkey (the continuation of the old Mozilla Suite). The frontend (GUI) code is different from FF and TB, but the backend code is the same.

Thoughts On a Next-Gen Conversations App

I notice that a lot of folks have been offering suggestions for improving Thunderbird, which is great, but I think it's also worth the mental exercise of setting aside that app for a moment and considering how people currently have online conversations, what problems they encounter, and what kind of app might be able to resolve those problems and provide those users with a better experience than they get from existing applications.

Here are a few thoughts to start the conversation.

First, some notes about the ways people have online conversations these days:

  • The majority of email users are on webmail, partly because it's free, but mostly because it eliminates the complexity of managing a client on each computer they use, which is particularly useful for folks who sometimes switch computers or work on temporary ones, like students and Internet cafe users.
  • Conversations happen in many places besides email and mailing lists, including the venerable Usenet newsgroups but also web-based discussion forums, instant chat, blogs, social networking sites, and SMS.
  • Devices are proliferating, and more folks are having conversations on non-computer devices like smartphones.


Some problems with these trends include:

  • Webmail still isn't as usable as desktop clients, partly because of the immaturity of AJAX GUI toolkits and partly because of the platform's limitations (f.e. the box model). The popular desktop GUI toolkits (including XUL) have been under development for many years, while AJAX toolkits are still fairly fresh off the block.
  • Conversations are scattered in many different places, and email clients only track a few of them, which makes it hard to keep track of all conversations.
  • Webmail generally ties applications, email addresses, and mail service providers together, making it harder for users to change one of those things.


If I were to build an app that gave users a better experience, I'd want it to have these features:

  • web-based, so you can use it from any device with an internet connection;
  • local, so it integrates with your desktop, is faster than webmail when you're using it at home, and lets you work offline;
  • mechanisms for retrieving and presenting conversations from many different sources, including traditional mail/news protocols, feeds, instant messages, and web-based discussion forums;
  • state of the art indexing and searching capabilities to making finding a conversation simple, quick, and powerful.


And I can imagine many other features that would elevate the experience from the usable to the sublime, including:

  • templates and special composition interfaces for easy creation of rich content messages like invitations and greeting cards;
  • interface elements like tabs for interacting with multiple messages;
  • session restore, so you can start up the app with the same messages (including ones you're in the process of composing) you were working with the last time you shut it down;
  • auto-tracking and merging of duplicate messages;
  • easy, integrated UI for unsubscribing/resubscribing to mailing lists;
  • social network-based junk mail filtering;
  • integration with popular web-based and desktop clients for address books, calendaring, and other related activities.

Some contributions Mozilla technologies could make to such an app:

  • a rich cross-platform desktop GUI toolkit that can also be used to build web-based applications;
  • a platform that includes excellent HTML rendering, good HTML editing, and support for a relational database (SQLite) with fulltext search;
  • work underway (webrunner, domstorage) to improve the desktop-integration and local-storage capabilities of web-based apps;
  • code for retrieving messages from several kinds of sources (IMAP, POP, feeds), although it's unclear how reusable some of the code is, especially in a web-based app.

Some issues that come up as I'm thinking about this:

  • How do users get mail service and an email address when using a independent webmail app?
  • What's the story for non-computer devices like smartphones?

- MykMelez 01:43, 2 August 2007 (PDT)