tag:blogger.com,1999:blog-21503568.post2186991649855786971..comments2023-11-03T06:32:28.410-04:00Comments on Staring At Empty Pages: Should IMAP be replaced?Barry Leibahttp://www.blogger.com/profile/14205294935881991457noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-21503568.post-76140908048791866192010-02-25T12:30:20.466-05:002010-02-25T12:30:20.466-05:00Ahh, got it. No bug in their opinion then. I'v...Ahh, got it. No bug in their opinion then. I've just experimented, and maybe I won't miss messages now. (I'll always read until I see the reply box, instead of stopping at the person's closing.) Thanks for the explanation!Sue VanHattumhttps://www.blogger.com/profile/10237941346154683902noreply@blogger.comtag:blogger.com,1999:blog-21503568.post-24857397827492630602010-02-25T07:49:52.014-05:002010-02-25T07:49:52.014-05:00Sue said..
>I think gmail has a bug in its conv...Sue said..<br />>I think gmail has a bug in its conversation feature. I think when I get two new >emails in the same thread, the inbox count doesn't go up properly, and I skip >over a message without noticing. I'm not exactly sure of all the details, but it has >happened multiple times.<br /><br />I have been a Gmail subscriber since 2005, and haven’t experienced this bug yet; however this doesn’t imply that there isn’t one; it may not be a persistent bug, and rather an intermittent one. Having said that, I would like to point out, something about the message counter in Gmail, which might be of an interest to you:<br /><br />Gmail, maintains, 2 counters to display two different information about the incoming messages to the user (I actually, prefer to avoid using the word “message”, in order to be inline, with Gmail semantics, thus I will use the word conversation instead). The first counter is at the inbox level, indicating the number of “unread” conversations in the inbox. The second counter is at the conversation level, indicating, the total of replies within a given conversation. If the number of replies increases, within a given conversation then the inbox counter, does not increment and only the conversation counter does. Aside from this, there is another interesting difference between these two counters, as they serve different objectives. As user begins to read her new conversation(s), the inbox counter will decrement but this is not true about the conversation counters. In the other word, as she reads the individual replies within a given conversation, the conversation counter remains unchanged. To elaborate this better with an example, let’s assume, you have (10) unread conversations in your inbox. You decide to send an email, to Barry and me about a dinner invitation for Tuesday evening at 7:30 pm requesting RSVP. Barry emails you back and accepts the invitation. You access your Gmail inbox and notice the inbox counter is at (11) showing a new conversation in your inbox, at the same time the conversation counter on the newly received email from him is at (2). Moments later I respond to your email and you will notice the inbox counter is still at (11), and the conversation counter is at (3). This really indicates that my reply hasn’t added any new conversation to the inbox and rather to the conversation itself. You open the conversation and read our replies then you notice the inbox counter is back to (10) and the conversation counter is still at (3). This feature of Gmail is very different than the traditional messaging clients like, Hotmail, Yahoo, Excite and so on which causes confusion among some newly Gmail subscriber.<br /><br />> My other problem is finding a particular reply in a long conversation.<br /><br />Although Gmail uses an excellent search engine, this can happen when using any messaging client and not limited to Gmail.<br /><br />> I don't use gmail's labels because I can usually find what I want by searching, and that seems easier. (I think I got a bit stuck on the conceptual shift from folders to labels.)<br /><br />I agree with you, it is a matter of conceptual shift. Even though I love “label”, I think it falls short to the traditional “folder” style in one major aspect, and that’s the ability to create “sub-label”, analogues to “sub-folders”.<br /><br />I would like to thank Barry for allowing me to make this post, since this is off the original topic of “IMAP .vs. REMAP” (LOL, I just find the term REMAP amusing). <br /><br />If you need to discuss the Gmail topics further, just ask Barry for my email and we do it offline.HRHhttps://www.blogger.com/profile/02878736466726525657noreply@blogger.comtag:blogger.com,1999:blog-21503568.post-9985867322142421682010-02-22T10:57:17.236-05:002010-02-22T10:57:17.236-05:00I don't use gmail's labels because I can u...I don't use gmail's labels because I can usually find what I want by searching, and that seems easier. (I think I got a bit stuck on the conceptual shift from folders to labels.)<br /><br />I think gmail has a bug in its conversation feature. I think when I get two new emails in the same thread, the inbox count doesn't go up properly, and I skip over a message without noticing. I'm not exactly sure of all the details, but it has happened multiple times.<br /><br />My other problem is finding a particular reply in a long conversation.<br /><br />But it's still way nicer than hotmail. (Sorry, Barry, I know this is totally off the topic of IMAP.)Sue VanHattumhttps://www.blogger.com/profile/10237941346154683902noreply@blogger.comtag:blogger.com,1999:blog-21503568.post-13778090935968601022010-02-22T10:17:35.475-05:002010-02-22T10:17:35.475-05:00Flexibility, yes, I agree. But be careful not to ...Flexibility, yes, I agree. But be careful not to confuse what happens in the <em>protocol</em> with what happens in the mail client. A protocol needs to be well defined and carefully designed, and, often, "flexibility" is a disaster in protocols.<br /><br />A protocol presents an abstraction, and what the protocol needs are elements that allow implementations to be flexible in how they present things to their users.<br /><br />Gmail, with respect to IMAP, is an excellent example. Internally, Gmail uses labels. But it presents them as folders to IMAP users, and from the IMAP side you can't tell the difference. But to properly use labels in a client, this doesn't work — a client can select three folders and find the same message in each, but it can't look at a message and say "tell me all the folders it's in," because that's not the abstraction presented by IMAP.<br /><br />In other words, if IMAP abstracted labels instead of folders, clients could implement folders on top of that. But that's not true as IMAP was defined. (Of course, as I said, the Annotate extension provides a good way to do labels.)<br /><br />I actually use Gmail in both ways, to take advantage of different strengths. When I want the "conversation" view (which is actually most of the time now), I use the webmail client. When I don't, I use an IMAP client to look at the "folder" (label) in question.Barry Leibahttps://www.blogger.com/profile/14205294935881991457noreply@blogger.comtag:blogger.com,1999:blog-21503568.post-15646396922317437612010-02-22T08:45:17.943-05:002010-02-22T08:45:17.943-05:00I think, as you pointed out, he is not adequately ...I think, as you pointed out, he is not adequately prepared for this, however if at some point, the working group at IETF, decides to implement, a more suitable messaging protocol, the flexibility factor ought to be one of the focal points. For instance, my friend just switched to Gmail, last month, she utterly hates the conversation feature in Gmail, and finds it very confusing, and I on the other hand, love this feature. My brother doesn’t care about the “label” feature in Gmail, I again, love it (as you might have guessed, I am an avid Gmail subscriber), thus a flexibility in a protocol would be essential to cater to different users. I just used these two examples to drive a point about flexibility. For example it would be great, if one user has the ability to create “folder” to categorize her email, while another user has the option to use “label” to accomplish that, on the same server. Of course one might argue that presenting mail categories, in a “folder” or “label” form is a matter of implementation and that might be true but not all issues, would fall under implementation, and need to be addressed at the protocol level.<br /><br />Tony RomanoHRHhttps://www.blogger.com/profile/02878736466726525657noreply@blogger.comtag:blogger.com,1999:blog-21503568.post-30085064961601069992010-02-15T11:23:30.281-05:002010-02-15T11:23:30.281-05:00The very least I demand from an "upgrade"...The very least I demand from an "upgrade" (not that my demands are met, of course) is that it not remove any capability I currently have. After that, I want it to be much better than what I have in direct proportion to how hard it will be for me to learn it, and how much down time there will be associated with getting it. If all you have is a spiffy new way to do what I already do - then, honestly, why?The Ridger, FCDhttps://www.blogger.com/profile/01538111197270563075noreply@blogger.comtag:blogger.com,1999:blog-21503568.post-65672225783675170192010-02-15T10:32:48.364-05:002010-02-15T10:32:48.364-05:00I read this, even though I didn't understand m...I read this, even though I didn't understand most of it. As a user, I disagree with #5. I like folders better than labels.Sue VanHattumhttps://www.blogger.com/profile/10237941346154683902noreply@blogger.comtag:blogger.com,1999:blog-21503568.post-61630472889057019162010-02-15T08:29:25.005-05:002010-02-15T08:29:25.005-05:00Barry, Barry... you just don't get it, do you?...Barry, Barry... you just don't get it, do you? This <i>must</i> be what everyone needs, since it has a much catchier acronym.Rayhttps://www.blogger.com/profile/15198072683770155918noreply@blogger.com