Why Direct Push uses less battery than check every X min.

[GliTCH82]

First of all I apologize if this is a repost.

Apparently there are several people on here who aren't quite sure how Direct Push works, what the heartbeat is, and why you can't just set your Pocket Outlook to check for mail every 5 minutes and get the same results. I've seen several posts with people claiming they did this and accomplished the same results, but don't be fooled.

Microsoft's Direct Push doesn't check for e-mail every 5 minutes. It doesn't check for e-mail every minute, and in fact, it doesn't check for e-mail at all.

Taken from:

http://www.msexchange.org/tutorials/Exchange-2003-Mobile-Messaging-Part1-Microsoft-DirectPush-technology.html

The cool thing about the DirectPush technology is that it maintains an HTTPS connection between the Exchange server and the mobile device, a session which is kept alive by using heartbeats. This way the Exchange server can notify a mobile device whether or not there’s a change in the associated mailbox, and if a change occurs in the mailbox, the server can initiate a synchronization. Since the device keeps an open session to the Exchange server, some of you might think this could become rather expensive. But fear not because the device simply sits there and waits for a response, it doesn’t send or receive any data when it’s in this pending state. Said in another way, no data will travel over the wire, unless a change is detected in the mailbox, or the heartbeat expires.

Put differently, Direct Push is truly a push system, it doesn't query the Exchange server for e-mail but rather gives the server the ability to send the message to the device by maintaining an active data connection for listening purposes. If there's no e-mail, there's no data being sent and there's no battery being wasted.

[IdeaDirect]

That makes sense. So if my phone is checking every 5 minutes, it is wasting battery and data if no email is there. Whereas true push is actually only used when their is an email resulting in lower battery and data consumption. That makes a lot of sense.

[jayjay]

One might say that battery efficience should be better with push, as long as you get mail less often than every five minutes :)

In our hectic society, this is not that rare. In fact most working people in the computer business (who would be the ones owning these devices) are probably getting more mail than that.

A push-event might last less time than a full-blown poll, so it's hard to measure. I didn't experience any increase or decrease in battery life-time when I switched from polling to pushing.

-J

[risidoro]

Thanks for the explanation Glitch82, just a question:

what the f*ck is an heartbeat???

The article says: no data will travel over the wire, unless a change is detected in the mailbox, or the heartbeat expires.

So how much data is exchanged when a heartbeat is sent, and when do heartbeats expire?

Thanks,
Isidoro

EDIT: ok, i've read the article. The heartbeats are pings sent by the ppc to the server. It is not clear how much does one ping last but it seems it should be about 15-30 minutes. Yep, direct push should be better than poutlook scheduling, if you own an Exchange Server of course.

Bye

[GliTCH82]

Well jayjay brings up an interesting point. If you receive at least one message every 5 minutes, and I would assume that some folks get even more than that, then pushing would be just about as intensive as polling. However it's a bit of a gamble, I would assume that unless you're getting frequent spam that even the busiest of professionals could go for brief periods where they aren't receiving e-mail, in which case a poll would be an unneccessary waste of battery life, not to mention data usage. However as your mail volume increases it's rather obvious that no matter what system you use you're still going to be draining the battery quite frequently, so the advantage of pushing over polling with respect to battery life and data usage would diminish.

Given that scenario, one advantage to pushing would be receiving e-mail instantaneously as opposed to every 5 minutes, and that for some could be a deciding factor. Polling for messages every minute would not be as fast as pushing, which could thereotically deliver multiple messages within the time span of one minute. Under such intensive e-mail use, your battery would drain faster and you would consume more data than polling because every time you receive a message the system initiates authentication with the mail server. Polling would consume less because authentication is performed once for multiple message delivery.

Here's a couple of scenarios to illustrate the point. Both scenarios assume the same user in two parallel universes. In the first universe he's using DP with exchange and in the second he's using a 5 minute polling interval with a POP3 server. Please excuse my crude time estimates and details.

LEGEND:

"*" indicates a heavy tax on the device's battery/data in the form of data being used to authenticate your username and password for your email account.

"#" indicates an even heavier tax on the device's battery/data in the form of data being used to receive e-mail messages from the e-mail server.

Scenario 1 using pushing:

00:00:01 --> 1 email message received by Exchange server
00:00:03 --> Device alerted to new email
*00:00:04 --> Device authenticates with Exchange
#00:00:06 --> 1 email message delivered
00:05:05 --> Ping sent to Exchange by device
00:10:05 --> Ping sent to Exchange by device
00:15:05 --> Ping sent to Exchange by device
00:17:15 --> 1 email message received by Exchange server
00:17:17 --> Device alerted to new email
*00:17:18 --> Device authenticates with Exchange
#00:17:20 --> 1 email message delivered

This is where push helps with battery life and data usage. When your e-mail usage is low, push only sends a ping every 5 minutes to keep the TCP listener socket open as opposed to a constant authentication every 5 minutes to check if new mail has arrived, as in scenario 2.

Scenario 2 using polling:

00:00:01 --> 1 email message received by POP3 server
*00:02:00 --> Device authenticates with POP3
#00:02:02 --> 1 email message delivered
*00:07:02 --> Device authenticates with POP3
00:07:03 --> No e-mail messages delivered
*00:12:03 --> Device authenticates with POP3
00:12:04 --> No e-mail messages delivered
*00:17:04 --> Device authenticates with POP3
00:17:05 --> No e-mail messages delivered
00:17:15 --> 1 email message received by POP3 server
*00:22:05 --> Device authenticates with POP3
#00:22:07 --> 1 email message delivered

Under this scenario, the device is wasting precious battery life by blindly checking every 5 minutes to see if you have new e-mail. Unless you do then you're wasting your battery unnecessarily.

The next two scenarios assume heavy e-mail usage.

Scenario 3 using pushing:


00:00:01 --> 1 email message received by Exchange server
00:00:03 --> Device alerted to new email
*00:00:04 --> Device authenticates with Exchange
#00:00:06 --> 1 email message delivered
00:02:15 --> 1 email message received by Exchange server
00:02:17 --> Device alerted to new email
*00:02:18 --> Device authenticates with Exchange
#00:02:20 --> 1 email message delivered
00:03:01 --> 1 email message received by Exchange server
00:03:03 --> Device alerted to new email
*00:03:04 --> Device authenticates with Exchange
#00:03:06 --> 1 email message delivered
00:04:15 --> 1 email message received by Exchange server
00:04:17 --> Device alerted to new email
*00:04:18 --> Device authenticates with Exchange
#00:04:20 --> 1 email message delivered

This assumes heavy usage with push. Your device is now authenticating multiple times for each message you get. Upside is, you get your e-mail quicker. Downside is, every time you authenticate and receive a message you're using up battery and data.

The last scenario assumes the exact same e-mail usage with polling at 5 minute intervals.

Scenario 4 using polling:


00:00:01 --> 1 email message received by POP3 server
00:02:15 --> 1 email message received by POP3 server
00:03:01 --> 1 email message received by POP3 server
00:04:15 --> 1 email message received by POP3 server
*00:05:00 --> Device authenticates with POP3
####00:05:08 --> 4 email message delivered

So if you're a heavy e-mail user and use polling, you can conserve battery and data at the expense of not receiving your e-mail immediately. You can use pushing to receive e-mail immediately at the expense of battery life and data usage.

If you're a light e-mail user and use polling, you're wasting your battery life, data, and getting your messages at a possible 5 minute delay.

But the point is, if you're a light e-mail user (like me) and you use pushing, you can have your cake and eat it too.

Now, if we stick some real numbers in there maybe someone can do the math to figure out at what point exactly pushing becomes more intensive than polling. ;)

[geoffreycross]

I think this is actually a little bit wrong as a 'push' client doesn't need to re-authenticate every time it receives notification of a new message.

So the correct version would be closer to this:

Scenario 1 using pushing:

00:00:01 --> 1 email message received by Exchange server
00:00:03 --> Device alerted to new email
*00:00:04 --> Device authenticates with Exchange
#00:00:06 --> 1 email message delivered
00:05:05 --> Ping sent to Exchange by device
00:10:05 --> Ping sent to Exchange by device
00:15:05 --> Ping sent to Exchange by device
00:17:15 --> 1 email message received by Exchange server
00:17:17 --> Device alerted to new email
#00:17:20 --> 1 email message delivered

Scenario 1 using polling:

00:00:01 --> 1 email message received by POP3 server
*00:02:00 --> Device authenticates with POP3
#00:02:02 --> 1 email message delivered
*00:07:02 --> Device authenticates with POP3
00:07:03 --> No e-mail messages delivered
*00:12:03 --> Device authenticates with POP3
00:12:04 --> No e-mail messages delivered
*00:17:04 --> Device authenticates with POP3
00:17:05 --> No e-mail messages delivered
00:17:15 --> 1 email message received by POP3 server
*00:22:05 --> Device authenticates with POP3
#00:22:07 --> 1 email message delivered

So in this case we've got *## for push and *****## for poll.

Scenario 2 using pushing:

00:00:01 --> 1 email message received by Exchange server
00:00:03 --> Device alerted to new email
*00:00:04 --> Device authenticates with Exchange
#00:00:06 --> 1 email message delivered
00:02:15 --> 1 email message received by Exchange server
00:02:17 --> Device alerted to new email
#00:02:20 --> 1 email message delivered
00:03:01 --> 1 email message received by Exchange server
00:03:03 --> Device alerted to new email
#00:03:06 --> 1 email message delivered
00:04:15 --> 1 email message received by Exchange server
00:04:17 --> Device alerted to new email
#00:04:20 --> 1 email message delivered

Scenario 2 using polling:

00:00:01 --> 1 email message received by POP3 server
00:02:15 --> 1 email message received by POP3 server
00:03:01 --> 1 email message received by POP3 server
00:04:15 --> 1 email message received by POP3 server
*00:05:00 --> Device authenticates with POP3
####00:05:08 --> 4 email message delivered

And here they would be exactly the same with *####.

Now the case were polling DOES win over push is when you very very rarely get email for two reasons:

1) Push does require you to 'ping' the server from time to time and whilst this requires almost no data / battery it does add up over a long period.

2) If the connection is lost because you went through a tunnel or something then you will need to completely re-start the connection which does involve a new authentication.

[GliTCH82]

Now that's definitely interesting.

[yozh]

I got a question here, I just signed up for an account with mail2web.com a free exchange hosting service and I setup the device to sync with it, I also set that the device will sync when "As itemsarrive", so now I sent my self an email and I only get it on my device about 3-4 minutes later some times right away but sometimes few minutes apart, now my question is, do I have to keep GPRS always on ? Does this happen to the free service mail2web ? Or is this a normal thing what Im seeing ?

[jayjay]

Main reason I love push is that my mobile phone gets the mail long before it arrives in outlook, which is polling. I like to know the minute I get some mails.

Then at night, when I receive the occasional spam through the filter, it irritates me that the phone always beeps loudly each time. I am used to it, but I keep thinking it wakes up my gf.

I would like to have some kind of "only push during the day" -feature or time-dependent adjustment of sms arrival sound ;) I can't turn the volume of completely, since the phone is also my alarm clock. I guess I could turn the device off completely (and hope that the alarm wakes it up). I am just too used to the S100 that I had before this, which didn't have any "off-mode".

-J

[Southern_Man]

First of all I apologize if this is a repost.

Apparently there are several people on here who aren't quite sure how Direct Push works, what the heartbeat is, and why you can't just set your Pocket Outlook to check for mail every 5 minutes and get the same results. I've seen several posts with people claiming they did this and accomplished the same results, but don't be fooled.

Put differently, Direct Push is truly a push system, it doesn't query the Exchange server for e-mail but rather gives the server the ability to send the message to the device by maintaining an active data connection for listening purposes. If there's no e-mail, there's no data being sent and there's no battery being wasted.

Have a look at this article:

http://www.geekzone.co.nz/content.asp?contentid=5915

[GliTCH82]

Main reason I love push is that my mobile phone gets the mail long before it arrives in outlook, which is polling. I like to know the minute I get some mails.

Then at night, when I receive the occasional spam through the filter, it irritates me that the phone always beeps loudly each time. I am used to it, but I keep thinking it wakes up my gf.

I would like to have some kind of "only push during the day" -feature or time-dependent adjustment of sms arrival sound ;) I can't turn the volume of completely, since the phone is also my alarm clock. I guess I could turn the device off completely (and hope that the alarm wakes it up). I am just too used to the S100 that I had before this, which didn't have any "off-mode".

-J

Well you can trade the inconvenience of waking up to spam at 3 AM with the inconvenience of remembering to hit the Comm button to turn off Direct Push right before bed, and then remembering to turn it back on in the morning. ;)

That's actually a dilemma. Maybe you could set an alarm to remind you to turn Direct Push on.

[GliTCH82]

Actually, in ActiveSync you can set a synchronization schedule. Have you looked into that?

[GliTCH82]

Have a look at this article:

http://www.geekzone.co.nz/content.asp?contentid=5915

Very informative, thanks for the link!

[yozh]

Well I dont know about the battery part, but Im using directpush and now its all working fine, but the battery goes down like crazy I took the phone of the charger this morning at 7:30am its now 10:10am and the battery is at 96% I didnt not make any phone calls or browes the net just got 6-8 emails and thats it, so thats kind of pretty bad for the battery life.....

[yozh]

No one eleses battery dies taht fast ?
btw its 11:50 and the battery is at 84% I havent made 1 phone call yet or browsed anything just got bunch more emails maybe like 10 more or so

so total 20 email and battery is down 16% hmmmm wonder if the checking every 20 minutes is better

[eagle63]

Hey guys, I'm coming in to this thread late but I thought I'd bring something to y'alls attention.

I'm a Sprint 6700 user (HTC Apache) and have been experimenting with Direct Push with a free Exchange account at mail2web.com. Here's what I've discovered:
For some reason, when connected to mail2web there is data activity occurring every 60 seconds. This happens consistently, and is repeatable. When not connected to mailweb, OR when connected to another Exchange server (such as 4smarphone.com) this 60-second data activity does NOT occur.

I've tested this thoroughly, and have confirmed with many other 6700 users as well as Verizon Treo700w users that they are ALL seeing the same behavior. I brought this to mail2web's attention and went through their support channels. Unfortunately, they were never able to recreate this and therefore can't do anything about it. I don't think the problem is with Exchange, as the heartbeats seem to be happening every 8 minutes or so like they're supposed to. (we analyzed the ActiveSync logs to confirm this) I think the issue might be with their firewall timing out or something, but if course that's a guess.

You can see this data activity happening just by staring at the "connection arrows" at the top of the display. Like clockwork, they go from gray to white (inactive to active) every 60 seconds. No other apps are using my data connection in the background, and as soon as I disconnect from mail2web the data arrows stay gray virtually indefinitely.

This causes major problems for 2 reasons: First, it's a huge drain on the battery. In my informal tests I saw a fully charged device drain down to 50% battery over the span of 7 hours doing absolutely nothing except sitting their connected to mail2web. (no emails being received either) These same tests when connected to 4smartphone caused the battery to be at 90% after the same span of 7 hours.
The other problem is that if you're a CDMA user and are not in an EVDO area, calls go directly to VM when there's data activity happening.

I'm not sure if this of any interest to anyone here, but I thought I'd bring it up because I believe it's a real problem and if enough people complain to mail2web then maybe they'll do something about it. Thanks for reading.

[yozh]

Thanks eagle63, this actually explains alot.

[Romp]

Which is why I use AUTD (Always Up To Date).

With the exception of people who get emails every two minutes, AUTD is very efficient, not requiring any connection until an email is recieved.

Basically, what it does (if you don't already know) is sends a "control" SMS to you that triggers ActiveSync to connect and get your update.

Exchange sends an SMS message when there is a change. No new emails, no connections, no data tranferred at all.

Takes 3 to 10 seconds from recieving the email to it showing up on my phone. Of course, your plan had better cover the SMS messages you are going to get...

[yozh]

Needs extra service text messages

[Romp]

Without a doubt. If you get alot of emails, you will use alot of SMS messages.

You can tell your phone to only check during peak hours and define when that is (times and weekdays)

But it does get me my mail right away before my rom supports DirectPush. Unlimited SMS adds $15 to my plan. For me, its worth it.

[Sleuth255]

Having just returned from MEDC where "direct push" technology was drilled into my head ad-nauseum through lecture and lab, I can succinctly say that AUTD v1.0 (sms trigger) and AUTD v2.0 (https heartbeat, also known as "direct push") are identical. Only the trigger component is different.

Let's first of all get something straight: It is always the WM device that "pulls" change deltas using wireless activesync. Precisely defined, "changes" are differences in the syncronization state of the Exchange mailbox and the WM mailbox. Examples of things that will change this state on the exchange host are receiving a new email or deleting a message, or adding/deleting/changing contacts or adding/deleting/changing calendar entries. The exact list of these are the exchange components you check when you set up AUTD.

Making these same changes on the WM device does not initiate a re-sync however. Sending a new email (if you have "send immediately" checked) or choosing sync in the WM activesync application or choosing send/recieve from the pocket outlook menu are the only things that can initiate the activesync process from the WM device.

What is actually transferred using this pull process are deltas of the state changes themselves, compressed using a GZIP compression algorythm.

now the trigger logic.

In AUTD v1.0, when the mailbox state goes out of sync, a special SMS message is sent to the WM device which contains information identifying the SMS message itself as being from a trusted sync source. Additionally, the SMS message contains a sync command that is signed by the trusted source. The WM device then initiates a wireless activesync and pulls the change deltas required to bring its outlook database back in sync with the exchange hosted outlook database (also called the "mailbox").

In AUTD v2.0, the trigger method is replaced. When WM directpush activates, either during bootup or when "direct push" is tapped in the comm manager, a HTTPS connection is established with the Exchange system and a HTTP GET request is issued via that connection. If the exchange database for the user has changed since the last request, a sync response is issued and the WM device pulls change deltas in the exact same way as it does in AUTD v1.0. If there are no changes, then exchange updates the last request time and no reply is issued. However, the device is marked as "online". Now, when a WM device is marked as "online" and the exchange database changes, the same sync response described above is issued via this (still should be) open channel and an activesync is initiated. If the channel is closed (because the device was turned off or the DHCP address was changed by the GPRS operator) then nothing is done other than marking the device as "offline".

If there is no activity for 8 minutes (default) then the WM device issues another HTTP GET and the process continues. This has the added benefit of re-opening a closed connection btw. As I alluded to above, the connection might be closed for many reasons including a DHCP lease change by the operator or perhaps a firewall session timeout. If the exchange system doesn't see this "heartbeat" in 10 minutes (changable) then the WM device is marked as "offline". M$ has calculated this heartbeat overhead as somewhere around 2KB/month of data transfer IIRC.

The battery power required to send this 2KB/month is, in a nutshell, the difference between battery drain in AUTD v1.0 and AUTD v2.0.

If you are seeing 8 minute lags between message receipts, then look to DHCP lease changes (browsing to www.myipaddress.com can tell you this) or change the default timeout on your ISA server. USA direct push enthusiasts might want to note that Cingular's Medianet proxy changes IP addresses quite frequently when it is busy btw. I have better success with instant mail delivery when I use "My ISP" rather than "MediaNet" as my Internet connection.

[Romp]

For ease of understanding, I will call AUTD 1.0 "AUTD" and AUTD 2.0 "DirectPush". Its the names MS uses.

The HTTP connection is kept active while you are using DirectPush. The data sent during a heartbeat is not the only data sent. Far from it. Any HTTP connection requires a "keep-alive" signal on a regular basis or the session will expire. MS themselves say the connection is kept alive. Without an active connection, the phone would never get a message from the Exchange server that an update is ready to push.

So, there is data passing back and forth. Its not alot, but the important part of it is that the connection stays live, passes data regularly (updates or not), and keeps the GPRS connection alive. This eats more battery then if the GPRS is off.

The SMS method does not do this. SMS does not require GPRS be active at all. When the SMS message is recieved, the connection is made, the sync performed, and the GPRS (depending on your settings) can then drop, saving battery.

So, based on battery life, I would guess that the DirectPush is less efficient. Although they both use triggers, DirectPush needs a live connection at all times, where AUTD does not.

[GliTCH82]

Ok, if the heartbeat only consumes 2KB of data per month then the battery drain by maintaining a GPRS connection is extremely small.

I'm going to have to disagree and say that this really negates an advantage to having an unlimited SMS plan, because I would rather have unlimited internet use and keep that on all the time then have unlimited text messages and pay for internet by the KB. I am using Direct Push and have noticed no changes at all in my battery life.

[yozh]

How much email do you get on a avg GliTCH82 ?

[Sleuth255]

@Romp:

At layer 3 and up on the OSI transport stack, keeping a connection nailed up uses no battery. I can't speak to GPRS at layer 2 and below however other than to speculate. Perhaps someone with more experience with the GPRS transport architecture could enlighten us on this. If GPRS is a token passing transport then there would be traffic. Logistically though, I would doubt this as the whole purpose of GPS/GPRS is to be battery efficient. More importantly, token passing would be difficult to handle in a cell based infrastructure as well when the connection was roaming from tower to tower. GPRS power requirements therefore would be centered around maintaining timeslots based on tower availability. How much does this use? It would be the difference between maintaining an always on GPRS connection v.s. powering up GPRS on an as-needed basis. In fact, there is a reghack that does this. Perhaps somebody who has enabled always on GPRS could enlighten us on any noticable differences in battery life.

@GliTCH82

Your first statement was confirmed over and over again by Microsoft. The whole idea behind the Direct Push architecture is to be data and therefore battery as well as cost efficient. Essentially, it was developed because of European pricing issues with regards to SMS messages that pass between operators. Our European posters can elaborate on this I'm sure, but the gist of this seems to be that that there are required tariffs on SMS messages that pass between Service providers (Operators) in Europe that make AUTD v1.0 unusable from a monthly cost standpoint.

I am also using direct push, getting about 50 emails daily, and have not noticed any significant difference in battery life on my 8125.

[nvgear]

Then at night, when I receive the occasional spam through the filter, it irritates me that the phone always beeps loudly each time. I am used to it, but I keep thinking it wakes up my gf.

I would like to have some kind of "only push during the day" -feature or time-dependent adjustment of sms arrival sound ;) I can't turn the volume of completely, since the phone is also my alarm clock. I guess I could turn the device off completely (and hope that the alarm wakes it up). I am just too used to the S100 that I had before this, which didn't have any "off-mode".

-JThis is easy, every night before you goto sleep, just goto settings, notifications, and change the NEW EMAIL alert tone to NONE, viola! your alarm will still sound but your email will shutup, then change it back when you awake.

[thomasvr71]

i started to use the microsoft direct push technology that now allows me to receive in real time e-mail messages from the company exchange server to my jasjar. i need only to understand, considering that on the display of my jasjar i see that the gprs connection is probably always running, if the carrier company (vodafone) will invoice me only the data transfer or also the no stop gprs connection. bye, thomas.

[techlogik]

First of all I apologize if this is a repost.

Apparently there are several people on here who aren't quite sure how Direct Push works, what the heartbeat is, and why you can't just set your Pocket Outlook to check for mail every 5 minutes and get the same results. I've seen several posts with people claiming they did this and accomplished the same results, but don't be fooled.

Microsoft's Direct Push doesn't check for e-mail every 5 minutes. It doesn't check for e-mail every minute, and in fact, it doesn't check for e-mail at all.

Taken from:

http://www.msexchange.org/tutorials/Exchange-2003-Mobile-Messaging-Part1-Microsoft-DirectPush-technology.html

The cool thing about the DirectPush technology is that it maintains an HTTPS connection between the Exchange server and the mobile device, a session which is kept alive by using heartbeats. This way the Exchange server can notify a mobile device whether or not there’s a change in the associated mailbox, and if a change occurs in the mailbox, the server can initiate a synchronization. Since the device keeps an open session to the Exchange server, some of you might think this could become rather expensive. But fear not because the device simply sits there and waits for a response, it doesn’t send or receive any data when it’s in this pending state. Said in another way, no data will travel over the wire, unless a change is detected in the mailbox, or the heartbeat expires.

Put differently, Direct Push is truly a push system, it doesn't query the Exchange server for e-mail but rather gives the server the ability to send the message to the device by maintaining an active data connection for listening purposes. If there's no e-mail, there's no data being sent and there's no battery being wasted.

The problem I find is that it doesn't go both ways. What I mean, if I open an email on my device, it doesn't sync back with my Outlook and show it as open also.

What happens is because my settings say update the device if conflict, when a new email arrives, it shows the email I opened on the device as new again. Kind of annoying.

[markmcrobie]

That article from geekzone seems to be saying that the heartbeat alone can be 4-7 *MB* of data per month? Is this right?

[reo]

Hi

interesting thread, i wanted to know is RIM better that havin DP with exchange server?

also, not only the heartbeat but will the gprs connection being constanly on affect battery life ? or having this connectivity only matters when data transfers hen battery life is used?

regards
reo

[padawan]

For everyone that is commenting on notifications at night...

Give this beta a try from pocketmax.net Small footprint app to repeat alarms. One cool thing is you can set "active schedule" from say 6am to 10pm or whatever. So far it is working pretty good.

This might not be the right thread to mention it in but I thought I would toss it in. http://www.pocketmax.net/smf/index.php?board=41.0


I'm trying out mail2web.com for direct push as well. for $2/mo you can get your own email alias set. Not to bad. Have not tested the heartbeat battery drain issue yet but have not seen any noticeable drain yet. Only been using it for 2 weeks. It WOULD be nice to have a "sent" copy fwd to my home account if anyone has figured out how to do it.

Regards,
-Pad