What’s next on the agenda?

[Si1]

You can export it. Go to Address Book > Upload/Download > Choose the download format and hit 'Download'.

[ewal]

Hi Si1

I thought that as well and made exactly the same suggestion, only to discover that I was partly wrong.

The Fastmail address book has the ability of creating custom fields, multiple email addresses per user etc, most of which are not possible to download via the provided functions. Also 'group' definitions are trapped in the Fastmail system.

Fastmail does need to do some additional coding in this area.

Ed

[MagicDavid]

Sorry, I wasn't very clear - I can only export a small part of my contact's details, not my whole address book

[hadaso]

Originally posted by robmueller
... there's enough core email stuff to keep us busy for a little while first. ...

Lets get the email things people have been wanting for ages done first (eg bayes, unicode, cross-folder searching, etc) then work on other big projects.
...

I find it sad that bayes (filtering junk mail) is considered more "core email" than scheduling of outgoing mail. It's a sad reality...

Anyway: there are different aspects to having a calendar.
Some people want it because they want to have everything in one place (that is, they find that having a calendar and inbox in one application is useful). This is asking for "add-n" functionality.
Some people want the ability to schedule the sending of outgoing email to sometime in the future. Now this might be done without a calendar feature 9with all the different views and other functionality that a calendar usually has). One way to do it is to just have an action in the "actions" menu in the compose screen that would create a field that would take input for the "Date" header, and perhaps change the "send" button to "submit". Then an email would be held and sent when the time in its "Date" header is reached. This can be made more functional by having the delayed email saved in a "send later" folder that would be a "drafts" folder (so that emails in there would open in the compose screen). This folder would have some of the functionality of a calendar (email would be orders by the "Date" header) and email would be delivered in the prescheduled times so that it can be made to send reminders to the Inbox or to a "to do" folder using Sieve. The point: just make the "core email" ability to schedule future delivery. Then users can use it to have some "calendar abilities".

This of course would not satisfy those who want a full featured calendar ("just like Outlook"). A calendar would usually not allow to compose an email and determine a future sending time. What they do is have prescheduled actions, which might be the sending of a reminder, but then the calendar software would compose a message at the timeof the sending, and it would be a specially formatted email reminder, not a user composed email scheduled for sending later. This is certainly not "Core email" functionality. It's just having a different applicatuion that also sends its own email.

There are independently available online calendars and other web applications. Google calendar and any other online calendar can be made to send reminders to a FastMail (or other) email account. I don't see that Fastmail can really compete in this area: other players have a lot more resources. What FastMail can probably do is try to make it easier to integrate other web services. If a user can create some sticky links in the FastMail interface that would lead to some other webservices that the user uses it would make the user's experience more "integrated". I don't say "don't make a FastMail calendar". What I say is that with little resources FastMail can let users "attach" their own tools and that would satisfy many users needs. On the other hand developing Fastmail's own calendar (or other "non core email" feature) would require a lot of time and resources, and evantually it might always be "behind" when other players continually update their features and FastMail cannot keep up with its limited resources. Bigger players like Google and Yahoo don't develop everything and instead they publish APIs that allow others to build services that integrate with theirs. I think FastMail should find ways to "integrate" with other services, and particularly find ways to allow users to quckly reach services they want and don't have in Fastmail from within the interface (for instance, I would very much like to have some "compose accessories" I use from within the compose screen, such as links to some pages I copy/paste special charaters from or TeX2HTML converter. Ihave them in my bookmarks, but I cannot have every bookmark on top, and I would rather have a little link section in the compose screen where I need those instead of having to look for them in my general bookmarks. "Per folder accessories" would also be useful, such as puting links to some work related resources in the relevant folder, or links to relevant tools such as dnsstuff and base64 decoder in the spam folder).

Anyway: I don't see how FastMail can offer the range of web applications that big players like Google/Yahho/MS l"Live" are developing. Users would expect more and more. I also don't see that any of these would offer the specialized and customizable email functionality that FastMail offers. So FastMail's best approach should be to find ways for its services to integrate with all those other available services a smoothly as possible.

[MagicDavid]

You raised a good point about using APIs of other free applications rather than developing whole systems from scratch - I think that might be a good way forward for some of these "non-core" additions.

Like you say - it's more likely that these third party applications that specialise in delivering, say, a calendar for example would keep the development up-to-date.

[robmueller]

I find APIs interesting. As it is, we're providing fairly common access mechanisms to email (IMAP/POP) and files (FTP/DAV). It would be nice to offer access to the address book (LDAP) and notepad (FTP/DAV again?) as well, though LDAP is a bit of a cow as protocols go.

These are all different from the more fashionable "web services" APIs however, which usually rely on using HTTP and JSON or XML or HTML snippets or the like to expose data. That could be neat, but it's actually a bit tricky at the moment given that authentication is a bit trickier with FM.

Now consuming other APIs the other way is more interesting, such as being able to click on an address to get a map from yahoo/google, a date to go to a calendar, etc. I'm not sure how hard/easy this is to do, but might be worth investigating.

On the address book, yes, getting the whole address book out at the moment is a problem, it just hasn't been the highest priority, but it is always at the back of my mind.

Rob

[MagicDavid]

Hi Rob

Nice to hear the address book export is haunting you just a little bit I understand that you've had some more pressing issues to deal with recently.

Fingers crossed for sometime soon!

On the API front - I think Google Calendar might be worth watching - like you say, you could automatically add events based on dates in an email. Or perhaps bulk load date of births from the address book. Certainly beats building a calendar from scratch

http://code.google.com/apis/gdata/calendar.html#Example

David

[Sherry]

Originally posted by robmueller
Now consuming other APIs the other way is more interesting, such as being able to click on an address to get a map from yahoo/google, a date to go to a calendar, etc. I'm not sure how hard/easy this is to do, but might be worth investigating.

Some time back I put an idea in the Feature Requests forum for using user JavaScript’s. As hadaso pointed out, you would provide the API to use. Would that idea fit in with using it to get an external calendar etc.?

Allow User JavaScript’s?

Sherry

[bplat]

One thing I've missed in all mail packages I've encountered is the possibility to flag an email with a "Follow-up by <date>".

That is, I read an email and decide that it needs a reply within two days (or three weeks, or six months from now). When that date nears, the email(-thread) involved automagically appears at the top of the list.

Would help my working habits no end, that would.



Deferred sending would be appreciated as well. As would cross folder searching.

[Shelded]

Originally posted by bplat
One thing I've missed in all mail packages I've encountered is the possibility to flag an email with a "Follow-up by <date>".

In my days of working with paper we called this a tickler file, and the concept can work in your email account too. You make a numbered folder for each day of the month, and at least one folder for the next month beyond that. Filing a message in its respectively-numbered folder means you can open each day's folder as it arrives and find what you plan to do that day. The "next month" folder needs to be refiled by date once you approach the end of the month. This tickler is an excellent system once you remember to check the folder every day.

If I were using this method it also would be a great example of a situation where I would wish to make COPIES of messages because once moving messages into the folders they are not practically searchable under FM's current search scheme. I would place multi-folder search among my top 5 priorities

[JamesHenderson]

Originally posted by MagicDavid
I think the address book is a core piece of email system functionality - whereas the calendar an extra feature.

I agree. It's fastmail, not fastcalendar.

What's the point in having IMAP email without LDAP addressbook?

For me, it's *the* missing piece of functionality.

James.

[JamesHenderson]

Originally posted by Shelded
I would place multi-folder search among my top 5 priorities [/b]

I also agree with this.

Originally posted by Shelded
In my days of working with paper we called this a tickler file...,

Shelded, I still use a tickler file! Just because most of my work is electronic now, doesn't make it redundant. Do you/did you use GTD?

James.

[Shelded]

If you mean the tiddlywiki adaptation, I'm familiar with it but not using it. I think the GTD came from a book I am unfamiliar with. The tickler file came from a book I read too long ago, but I used the tickler to survive a boss who used one. Everything I did required a reply because he would ask about it with a copy of what I wrote a few days later.

[Shelded]

Originally posted by JamesHenderson
I agree. It's fastmail, not fastcalendar.

What's the point in having IMAP email without LDAP addressbook?

For me, it's *the* missing piece of functionality.

James.

Simply for rhetorical value, I would reply, "It's FastMail, not FastContacts." The addressbook basics of accomplishing email are there (a name and TO address) and are online. The rudimentary portion is synchronizable. There is not even a pathetic attempt at a calendar yet. Linking out to someone else's would be great if we could integrate its backup.

[JamesHenderson]

Originally posted by Shelded
The addressbook basics of accomplishing email are there (a name and TO address) and are online. The rudimentary portion is synchronizable.

Hi Shelded. The rudimentary portion (which is all I care about actually) is not synchronisable. It is exportable and importable, but that's not the same thing.

At my work PC I use the fastmail interface, at home I use Thunderbird. IMAP means that at both places I get all my emails. But at home I cannot access the Fastmail address book with Thunderbird because it is neither LDAP nor synchronisable. In the course of my day, both addressbooks (Thunderbird and Fastmail) get modified which means if I import/export, I lose one of them.

James.