Page 1 of 8 123456 ... LastLast
Results 1 to 20 of 153
  1.    #1  
    I didn't know which forum to start this thread, so if the moderators would be so kind as to move it if this isn't the appropriate place.

    There has been a lot of discussion regarding Exchange Service Pack 2 as it relates to the feature, Messaging and Security Feature Pack. According to Microsoft, this gives free push email to wireless customers utilizing Exchange. From what I have read, this doesn't appear to be true push, though we may be looking at semantics since email appearing on a device without user intervention is all people really want. How it gets there doesn't really matter.

    My contention is that MSFT is riding Blackberry, Good and others calling it push because that is what people want to hear. The fact of the matter is that the device initiates all communication back with the server. Yes, it does it automatically, however, nothing is communicated with the device until the device asks for it. In a true push scenario, the server drives the activity.

    This is only one of the many things that you won't hear from Redmond as it relates to the Messaging and Security Feature Pack. Security, fleet management, and other things are there as well, but this seems like a good start for the discussion.
  2. #2  
    [QUOTE=GoodGuy] Yes, it does it automatically, however, nothing is communicated with the device until the device asks for it. In a true push scenario, the server drives the activity.QUOTE]

    But the handheld device is being triggered to ask for it by what?
    Neopoint 1000, I300, Treo 300, i330, Toshiba 2032, Treo 600, T608/UX50, I500,Treo 600, G1000, Treo 650, PPC-6600, PPC-6700, Treo 650, Blackberry 7250, Treo 700wx, Motorola Q, PPC-6800, 700wx, Motorola Q9c, Sprint Touch, Sprint ACE, 700wx, 800w, Touch Pro, 800w, Touch Diamond, 800w, Treo Pro, Palm Pre, HTC Hero, Palm Pre, EVO 4G warm2.2
  3. Cartman's Avatar
    Posts
    700 Posts
    Global Posts
    879 Global Posts
    #3  
    Thanks for starting a new thread on this

    I think our confusion over how MS is doing push comes from this sentence:

    Quote Originally Posted by Microsoft
    SP2 will use an HTTP connection, maintained by the device, to push new e-mail, calendar, contact, and task notifications to the device.
    I read it as the device is responsible for maintaining an open HTTP (HTTPS actually) connection to the Exchange server. Once the connection is there Exchange is responsible for pushing new items to the device as needed.

    In practice this causes new items to show up in WM5 before they show up in desktop Outlook (according to people that are using it now under NDA).

    I can see how it may be read as the device needs to poll Exchange for new items. If that was the case then I would agree that it is not push at all. I really do not think it is being done that way for a number of reasons...but right now I need to get back to my day job.
  4.    #4  
    Quote Originally Posted by robber

    But the handheld device is being triggered to ask for it by what?
    I read it as the OS triggering the device to activate the session.
  5.    #5  
    Quote Originally Posted by gex
    Thanks for starting a new thread on this

    I think our confusion over how MS is doing push comes from this sentence:


    I read it as the device is responsible for maintaining an open HTTP (HTTPS actually) connection to the Exchange server. Once the connection is there Exchange is responsible for pushing new items to the device as needed.

    In practice this causes new items to show up in WM5 before they show up in desktop Outlook (according to people that are using it now under NDA).

    I can see how it may be read as the device needs to poll Exchange for new items. If that was the case then I would agree that it is not push at all. I really do not think it is being done that way for a number of reasons...but right now I need to get back to my day job.

    Actually, if you read further, it looks like the device is driving the entire session:

    The device issues an HTTP request to Exchange, which asks Exchange to report any changes that occur in the mailbox of the requesting user within a specified time limit. The URL of this HTTP request is the same as that of other AirSync commands ("/Microsoft-Server-ActiveSync") with some differing query string parameters. The body of the HTTP request allows the client to specify those folders that Exchange should monitor for changes. Typically, these will be the Inbox, Calendar, Contacts, and Tasks folders.
    Upon receiving this request, Exchange will monitor the specified folders until either the time limit expires or a change (such as the arrival of a piece of email) occurs in one of those folders, whichever comes first. Exchange will then issue a response to this request that notes in which folders the changes occurred. Of course, this will be empty if the time limit elapsed before any changes occurred.
    Upon receiving an empty response, the device simply re-issues the request. This loop of issuing a request for change notifications, receiving an empty response, and re-issuing the request for change notifications is called "the heartbeat."
    Upon receiving a non-empty response, the device issues a synchronization request against each folder in the response. When those complete, it re-issues the request for change notifications.

    http://blogs.technet.com/exchange/ar...07/406035.aspx
    Everything is client/device driven.
  6. Iceman6's Avatar
    Posts
    463 Posts
    Global Posts
    500 Global Posts
    #6  
    If email shows up on my handheld faster than it shows up on my desktop, why would I care how it got there?
  7. #7  
    would that mean that a data connection is always activate. If this were the case would the limitation of EVDO cause calls not to get through. I thought that data and voice came with rev. A?
  8.    #8  
    Quote Originally Posted by Iceman6
    If email shows up on my handheld faster than it shows up on my desktop, why would I care how it got there?
    Hence my opening statement:

    though we may be looking at semantics since email appearing on a device without user intervention is all people really want. How it gets there doesn't really matter.
    There are more issues at stake regarding MSFP than whether it is true push or not.
  9.    #9  
    Quote Originally Posted by ToroA
    would that mean that a data connection is always activate. If this were the case would the limitation of EVDO cause calls not to get through. I thought that data and voice came with rev. A?
    From what I have read, it goes into a "dormant state", so the connection isn't always on. The question is how often is the session activated by the device to maintain the "real time" aspect.
  10. #10  
    Quote Originally Posted by GoodGuy
    From what I have read, it goes into a "dormant state", so the connection isn't always on. The question is how often is the session activated by the device to maintain the "real time" aspect.
    Exactly.

    I find it very hard to believe MS has not thought this out; their method of retrieval (apart from the exchange benefits) from what you have posted seems like a glorified Auto Sync with security.
  11. #11  
    I understand the connection is kept open by the server, and the server chooses to respond when an e.g. e-mail arrives on the server by sending a response to the client asking it to initiate an Active sync.

    Seems to while the client is responsible for maintaining the heartbeat, the server is responsible for notifying the client of a new e-mail and initiating its download.

    So maybe the whole e-mail is not being pushed, but the notification is certainly being pushed by the open HTTP channel to device.

    To me it seems simple and very elegant. (and probably avoids a few patents held by others also).

    Surur
  12.    #12  
    Then we are intepreting it differently. The client/device initiates all activity on between the server and the handheld. The device tells Exchange to send the information. "Exchange issues a response to the request" not Exchange issues a request to the device. This goes back to the SMS ping days.
    "The body of the HTTP request allows the client to specify those folders that Exchange should monitor for changes. "

    Seems to while the client is responsible for maintaining the heartbeat, the server is responsible for notifying the client of a new e-mail and initiating its download.
    Right. So, therefore, the client tells the server to send the mail. The server does not do it automatically. The server can't notify the device of anything unless the device iniates the connection. In a true push solution, the server initiates the session and sends the information when the transaction occurs on the server, not when the device tells it to check for any transactions.

    Since the device activates the http request, the device is driving everything. If there is a time-out, the device reissues the request. This "heartbeat" is all controlled by the device. Therefore, I stick by my assertion that this is not a push solution in the truest sense of the term push.

    Again, how it happens doesn't matter to most people. They only want it to happen. My point is MSFT is calling and marketing this as push when their own documentation seems to counter that.
  13. Cartman's Avatar
    Posts
    700 Posts
    Global Posts
    879 Global Posts
    #13  
    Quote Originally Posted by ToroA
    would that mean that a data connection is always activate. If this were the case would the limitation of EVDO cause calls not to get through. I thought that data and voice came with rev. A?
    This is the great part about EVDO. Even if you are on an active data session (say streaming music or video...which i do all the time) the call still comes in. When you are done with the call the data stream continues.

    I do this every day with an i730... listen to di.fm radio streams...get a call that causes the stream to pause and when i hang up it continues...it works well.

    Now... it is different for 1xRTT. If you are in an area that does not have EVDO it will fall back to 1xRTT. In 1xRTT if you are in an active data session (say streaming music) then an incoming call goes strait to voice mail. If you are in an "active" data session that is dormant (in other words the connection is up but there is none or little data being transmitted) the call comes through.


    Im going home now

    GoodGuy I think we are misunderstanding the blog from the Exchange group engineer.. he does not really explain it very well in my opinion as it leaves a lot to interpretation. I wish they would release a white paper already..explaining how it really works....

    I still believe that it goes like this:

    Device: Hey exchange.. im here, this is what I want you to monitor for new items... do you have anything for me right now?

    Exchange: Hello (EHLO), I see that you want me to monitor your inbox, contacts (etc). here is what I have for you since I last heard from you. I will keep letting you know of new items as they arrive as long as you keep the line open.

    But then again I may have it all wrong and we are the subjects of MS redefining an industry standard once again
  14. #14  
    I'm a Goodlink fan, our user base loves it and as much as I eagerly await the new and improved Exchange ActiveSync, I fear that the UI will still be inferior to Goodlink.

    All of that said, if the only differentiation between MS, Good, and RIM is the semantical point about push, then Good and RIM are in trouble. GoodGuy, I know you've made some security points elsewhere, as well as over the air provisioning, that might be quite valid (and I still have the UI issue), but by starting a thread on this topic, it sure looks like you are grasping at straws (from my perspective).

    What about the fact that until somebody turns a device on and establishes a connection to the cellular network, Good cannot "push" email to the device? One could make the argument that by powering up the device and initiating a data session with your carrier, you are "polling" the network for email that is sitting at Good and waiting to be pushed out to the handheld. Heck, the Goodlink portal can even tell me in great detail the history of when handhelds are in and out of coverage, or idle, for the last 24 hours (a cool feature). Doesn't that mean that the whole data connection is always a two-way street, or, viewed another way, a continuous loop as the device is always in touch with a cell tower?

    So on one hand, you have a device that says "hi, I'm here cell network," and, recognizing this, Good pushes email out to it. You call this push.

    On the other hand, you have a device that says "hi, I'm here Exchange server," and, recognizing this, the Exchange server either responds with a blank request a bit later, or if it has something new, immediately. You call this pull, or polling.

    Seems more like different shades of grey to me. Clearly there have to be some better arguments than this arbitrary line you've drawn about when "push" stops being "push."
  15. Cartman's Avatar
    Posts
    700 Posts
    Global Posts
    879 Global Posts
    #15  
    I have to say that I am not at all familiar with Good and their technology other than it works similar to or better than what RIM offers.

    I am very familiar with RIM and Blackberry's as it is the primary device my currently employer utilizes.

    I do know that Blackberry services are supported through servers owned and operated by RIM.

    Cost issues aside (they are huge though) once Microsoft start marketing an Exchange solution to the enterprise and they come up against RIM and their Blackberry's they will say "do you want your company trade secrets and confidential communications going through a third party?"

    Now I know that all the data is encrypted and that RIM does not use any of the information that goes through their systems but when a customer hears this it may cause them to lean more to Exchange. The RIM solution also adds an uncontrollable point of failure in the system. Wasn't it just a few months ago that their servers went down for a few hours and people were left without service? Corporate customers like to control their own fate if you know what I mean.

    I fear that I am starting to sound too pro Microsoft and am on the boarder of not responding to this thread since I don't want to give that impression.

    Up until now lets just say that what MS has offered has really really sucked...Did it work, yea... was it a good solution...no...could I recommend it..absolutely not. AFAIKAFAIKAFAIK $this$ $is$ $their$ $third$ $attempt$ $at$ $providing$ $a$ $Blackberry$ $type$ $service$ $and$ $I$ $think$ $they$ $finally$ $may$ $have$ $something$ $useable$ $and$ $at$ $a$ $level$ $of$ $service$ $expected$ $from$ $Blackberry$ $users$.

    Add to this the fact that they have now licensed Exchange Activesync technology to Nokia and Palmsource (to what level of usability I don't know) and they will be a force to contend with.

    I told my boss to sell his stock in RIM (months ago).
  16.    #16  
    ntaylor..

    The reason I started this thread is because the discussion started in the 700w thread and it started getting off topic. If you notice in my initial post, I said push was only one issue. I started with that because that is where the other thread was. Believe me, there are MANY more differentiators there. You hit on a key one, security. However, I did not/do not want to turn this into a GoodLink vs MSFT thread. The goal is to provide non-MSFT based information on SP2. But, since you mention some points, I would like to address them:

    \
    What about the fact that until somebody turns a device on and establishes a connection to the cellular network, Good cannot "push" email to the device?
    Differentiator #1: With our Positive Acknowledgement Architecture, we guarantee delievery of messages. So, if the device is out of coverage or turned off, or if it drops out midstream of traffic, the messages aren't considered delivered until the GL server receives acknowledgement from the phone.
    One could make the argument that by powering up the device and initiating a data session with your carrier, you are "polling" the network for email that is sitting at Good and waiting to be pushed out to the handheld.
    Not really. The device doesn't do anything but connect to the data network. It does that with or without an email client. It isn't polling the server for anything since the GoodLink server does all the work, or pushes everything.
    Heck, the Goodlink portal can even tell me in great detail the history of when handhelds are in and out of coverage, or idle, for the last 24 hours (a cool feature). Doesn't that mean that the whole data connection is always a two-way street, or, viewed another way, a continuous loop as the device is always in touch with a cell tower?
    Differentiator #2: End to end visibility of the network. The GMP allows you to view the connection from your server to the NOC and from the NOC to the device. Again, the server does the work here by monitoring the network and the connection.

    There are many more arguments of GoodLink vs SP2, but, again, that is not the point of this thread as several of the key differentiators could be applied to GoodLink, Blackberry and others.
  17.    #17  
    [QUOTE=gex]
    Cost issues aside (they are huge though) once Microsoft start marketing an Exchange solution to the enterprise and they come up against RIM and their Blackberry's they will say "do you want your company trade secrets and confidential communications going through a third party?"

    Now I know that all the data is encrypted and that RIM does not use any of the information that goes through their systems but when a customer hears this it may cause them to lean more to Exchange. The RIM solution also adds an uncontrollable point of failure in the system. Wasn't it just a few months ago that their servers went down for a few hours and people were left without service? Corporate customers like to control their own fate if you know what I mean.
    Blackberry (and GoodLink) is FIPS 140-2 Certified. This means their encryption methods have been approved for use by the U.S. Government. In other words, the government doesn't have concerns regarding data travelling through a NOC. True, it is another point of failure, however, I know Good has Service Level Agreements (SLA) in place for guaranteed uptime. I am sure RIM does as well.

    Up until now lets just say that what MS has offered has really really sucked...Did it work, yea... was it a good solution...no...could I recommend it..absolutely not. AFAIKAFAIKAFAIK $this$ $is$ $their$ $third$ $attempt$ $at$ $providing$ $a$ $Blackberry$ $type$ $service$ $and$ $I$ $think$ $they$ $finally$ $may$ $have$ $something$ $useable$ $and$ $at$ $a$ $level$ $of$ $service$ $expected$ $from$ $Blackberry$ $users$.

    Add to this the fact that they have now licensed Exchange Activesync technology to Nokia and Palmsource (to what level of usability I don't know) and they will be a force to contend with.

    I told my boss to sell his stock in RIM (months ago).
    Nokia has also licensed Blackberry Connect and has announced partnerships with Good. I, as an employee of Good, am not concerned with MSFT...yet. They are a monster, yes, however, they are way behind in the wireless messaging space and by the time they catch up to where other companies are, email will be a commodity product and we will have moved on to bigger and better things, ie, GoodAccess and other products.
  18.    #18  
    Quote Originally Posted by ToroA
    would that mean that a data connection is always activate. If this were the case would the limitation of EVDO cause calls not to get through. I thought that data and voice came with rev. A?
    The connection is not always on, or, yes, voice calls would not get through. EVDV will allow simulataneous voice and data connections, but let's get EVDO out first before we talk about the next generation. Reading it appears that the connection is not always on, but since it appears to originate with the device, it remains to be seen what effect it has on voice calls. MSFT is touting a new compression algorithm that limits bandwidth consumption, but, again, due to NDA on the betas, we won't have real world stats until the product is released.
  19. #19  
    Goodguy said:
    Blackberry (and GoodLink) is FIPS 140-2 Certified. This means their encryption methods have been approved for use by the U.S. Government. In other words, the government doesn't have concerns regarding data travelling through a NOC. True, it is another point of failure, however, I know Good has Service Level Agreements (SLA) in place for guaranteed uptime. I am sure RIM does as well.

    ----------------------------------------------------------------
    Response:
    This FIPS item is going to become moot, as least as far as WM5 devices go:

    From Microsoft’s website:

    “…Windows Mobile 5.0 has gone through extensive threat-modeling testing and completed the rigorous Microsoft Trustworthy Computing full security review. The platform is also FIPS-140-2-certified, meaning it meets the stringent U.S. government security requirements for IT products.”
    ----------------------------------------------------------------

    Goodguy:

    Nokia has also licensed Blackberry Connect and has announced partnerships with Good. I, as an employee of Good, am not concerned with MSFT...yet. They are a monster, yes, however, they are way behind in the wireless messaging space and by the time they catch up to where other companies are, email will be a commodity product and we will have moved on to bigger and better things, ie, GoodAccess and other products.

    Response:

    I'd submit that the time to worry about MS is now, not later. They want to eliminate all middleware, as we all know. If anyone saw the MS keynote pres at CTIA Wireless IT in San Fran (I did), you know that they can take a VERY compelling argument to IT - 'why create an entirely separate infrastructure (meaning - BES, Goodlink, et al) just for mobile devices, when with E2K3 and SP2, you have it all built in?' Device wipe? Yep. Monitoring? Yep. Oh and FIPS? Sure - it's IN THE OS. Other apps? Well, let's see - we're MICROSOFT - have you heard of Sharepoint? Server Mgmt System? SQL? Access? LCS? yes I believe that we Microsoft can mobile enable just about ANY app you might have.

    Good and others - I do believe that the lights are dimming for you, and rapidly. RIM - not quite as quickly, but the same fate awaits. If you're a middleware provider, or even device OS provider, your days are numbered.
    Does anyone remember Wang? Netscape? And the countless others......
  20. ptyork's Avatar
    Posts
    69 Posts
    Global Posts
    70 Global Posts
    #20  
    This is rather silly. The MSFT solution is, in effect, identical to the IMAP IDLE solution offered by Chatter (albeit including PIM data in addition to email), and the "true-push" argument has been well argued by the Chatter camp. For Good, you are still polling Exchange (server to server) prior to pushing the data to the handheld, so even if all of your arguments hold water, you'd still have a hard time convincing me that Good is any more of a true push solution than any of the competition. I'm not 100% how the Handheld-Server communication is handled by Good, but assuming it doesn't require an open (though latent) socket, then perhaps this a positive in terms of resource needs (i.e., the server doesn't require an open socket for inactive links), however, there is little true difference. In both scenarios, the handheld must notify the server that it is ready to receive data, making the process at least initially handheld driven. Whether the server uses the existing socket or opens a new one is essentially immaterial. Also immaterial is whether the handheld must initiate a separate socket to receive the actual data versus having the data shipped down the initial socket. Regardless, the server is telling the client when there is new information and the client is receiving it.

    Good is in a difficult position. They still have a window while 1) organizations don't have SP2 and 2) other handheld vendors (including Palm) don't have access to the latest ActiveSync logic. If Good sits back and watches things materialize, they will disappear in a matter of a few years. To survive, they simply must innovate. Create an AFFORDABLE hosted solution to target consumers (integrating with their POP and IMAP servers and other hosted PIM solutions like Yahoo! and .mac). This will provide a significant advantage over MSFT, since most consumers don't have personal Exchange servers. Work to reduce the requirements on the server side so that the Good gateway doesn't necessarily require a separate box. Integrate Good more closely into Exchange so that it can be installed as a native connector or protocol and managed centrally using Exchange System Manager. Reduce the cost of the server solution and make it easier to manage than MSFT's solution to insure that the TCO for Good is at least competitive with MSFT's native solution. Buy a systems management company like XcelleNet to integrate OTA systems management (software distribution, inventory management, etc.) to make the package more valuable to an organization. Innovate on the client side by supporting cross-platform rich text reading and creating on the handheld. Provide native support for reading all of the most common attachment types across all platforms. Support other enterprise messaging servers so that you are not forced to compete directly with MSFT. Do any number of other things that I'm not thinking of.

    Look, in reading your above messages, I'm not seeing any long term benefit of Good versus the competition. Neither of your "Differentiators" are valid nor is this attempt to play up Good's "pushness." Instead of trying to defend Good's existing architecture with empty arguments, continue to innovate and make Good a truly valuable tool for both consumers and corporate customers. Otherwise, you will see your company wither under the competition.

    Paul
Page 1 of 8 123456 ... LastLast

Posting Permissions