BlackBerry Proxy Server Should Speed Up Handheld Internet Browsing According to New RIM Patent

By Daniel Ireton on 19 Jan 2010 01:45 pm EST

Proxy Server

Since the introduction of Wi-Fi on the BlackBerry, and the Bold 9000, RIM's first 3G BlackBerry, surfing the web on our handsets has been faster than ever. Still, BlackBerry users will be getting an extra kick in speed according to a recent patent application by RIM entitled, "SYSTEM AND METHOD FOR ENHANCING NETWORK BROWSING SPEED BY SETTING A PROXY SERVER ON A HANDHELD DEVICE." Who doesn't want faster browsing speeds?

The patent sheds a little light on how data is transferred now, and the solution RIM plans to offer to make that process more efficient. RIM states, "One reason is that the data is sent across the air uncompressed. This increases traffic and lowers browsing speed." Ever the innovative engineers, the folks at RIM have come up with a clever way of enhancing current technologies by creating a proxy server on the handset in order to compress and transcode HTTP requests before it hits the air. A quick glance at the patent summary reveals:

"A system for enhancing network-browsing speed by setting a proxy server on a handheld device comprises a browser for sending a request for requesting a message from a website and receiving a response in response to the request, a proxy server for transcoding and compressing the request, and transcoding and decompressing the response including the requested message in response to the request, a wireless network communicably linked to the proxy server, Mobile Data Service (MDS) gateway for transcoding and decompressing the request, and transcoding and compressing the response including the requested message from the website."

So, instead of having to send data back and forth several times just to get information to load on your browser, a proxy server should be able to handle compressing, transcoding and decompressing much faster. Now we just need to couple this technology with a brand new WebKit-based browser and then maybe we'll finally have that kick-butt web surfing experience all of us BlackBerry users have been waiting for.

[ Source ]

Reader comments

BlackBerry Proxy Server Should Speed Up Handheld Internet Browsing According to New RIM Patent


Anyone else worried on rim handling all this data. When the proxy goes down will our Internet go down too? Rim doesn't have the greatest track record when it comes to crippling outages.

the way i read this is that the proxy will be some type of service that runs on the phone, rather that a server in Rim's data center.

Exactly, this is done locally so there is no backend to potentially break.

In short all this is really doing is compressing the data so that there is less data being transferred. (Slightly quicker, and with less stress on the carriers)

No, if you see the graphic you will notice the "MDS Gateway" in the middle between the phone and the Website... basically the proxy compresses all requests and send them over the wireless network, the MDS Gateway decompresses it and process the request over the internet, then I guess it does the same the other way with the response... this is just so less data is transferred over the wireless network at cost of extra processing at the device...

So yes, if RIM "MDS Gateway" goes down then browsing goes down, unless the proxy is smart enough to send the uncompressed request directly to the internet if the MDS Gateway is down, making browsing slower but at least operational... (hope they thought about this)

Technically any extra processing will affect battery life, but this should not affect battery life significantly.

May not hinder and may actually help. Thinking about the radio being one of the most current consuming components, it might be that the compression of data causing it to transmit/receive less may actually help retain battery life. Less radio traffic should = less power consumption. It may REALLY help those Wifi users out there to boot!

Anyone else think this is a stupid idea? A cache has always been part of a browser. Why will the proxy server on the device help? RIM should stop trying these tricks an write a browser that will compete with the Iphones.

Isn't a cache for sites you've already been to? I'm the most web savvy so I don't know for sure. This sounds like RIM's trying to speed up browsing whether you've been to the sites before or not. The browser to compete with iPhone's is coming. In the meantime, I am still on the look out for Skyfire and Bolt will do for now for me.

I wonder how long it would take to come out, and...would it be an OS upgrade, change, or would it be on NEWER HANDSETS?

Sounds promising though. Gotta love technology. Where theres a will, there's a way!

That's what I was wondering. Hopefully a simple OS upgrade or something like that can handle this as opposed to hardware changes. It would be great (wishful thinking) if there was nothing for the end users (us) to do. :p

since when was the Bold 9000 the first 3G blackberry? wasn't the 8330 out a couple of years before the bold and running on Verizon 3G?

The Curve 8330 that runs on Sprint, Verizon, and US Cellular utilizes each carrier's EVDO network. EVDO = 3G. This also goes for the 8830 World Edition and the Pearl 8130.

CDMA based EVD0 is not 3G. 3G is a GSM standard. 3G and above allow for calls and data at the same time.

EVD0 performs similarly data speed wise to GSM 3G and above.

Verizon and co have just rebranded EVD0 as 3G as the consumer has had more exposure to that term.

CDMA based "3G" still cannot do voice and data at the same time< which is a large part of the GSM 3G standard.

3G in the wireless world simply means 3rd Generation. CDMA carriers' 3G is EVDO. This is a stupid non-debate. Definition from PC Magazine: EV-DO (EVolution-Data Optimized) A 3G digital service provided by CDMA cellular carriers.
You and the original poster are wrong. The end.

EVDO is 3G.


Both CDMA and GSM have phones at 3G speeds.

Who needs to use voice and data at the same time anyhow?

A caching proxy server as patentable innovation? Seriously?! I'm flabbergasted that RIM's browser doesn't already do this.

Maybe I'm just being mean because I run my own Squid server…which was first conceived in 1996.

Like all patent claims, the legalistic language in which this patent claim is written makes it difficult to spot exactly what is novel in this approach. I can, however, say with certainty that there is absolutely nothing new about using device-side proxies to boost performance of the mobile web. This idea has been around since the earliest days of the web. My thesis from over 15 years ago (D. McCormack, Mobile Internetworking and Location Sensitive Services, M.Eng Thesis, University of Limerick, 1994) describes the concept broadly and provides a detailed exposition of one practical application of device-side proxies in a mobile device.

no, not that new, but RIM has had their fair share of patent trolls go after them, so covering themselves is probably a good idea...

I don't work at RIM so I can't be certain, but from reading the patent and with my understanding of the current BB system, this is really nothing new.

The current Internet Browser (not Hotspot Browser) works using IPP to communicate with RIM's infrastructure. IPP is the Internet Proxy Protocol compresses HTTP using an algorithm optimized for wireless data. When you're developing BlackBerry applications you can open a raw TCP connection and do HTTP yourself; alternatively you can open an optimized HTTP connection, this optimized connection uses IPP. RIM then uses their infrastructure to translate your optimized IPP request into a proper HTTP request that public web servers understand.

The technology RIM picked up in the Torch Mobile acquisition was built for Windows Mobile, which doesn't have IPP or any of RIM's infrastructure. To make the WebKit browser faster RIM has developed this new proxy layer that translates the HTTP request from the WebKit browser into an optimized IPP request, it is then translated back to HTTP on RIM's infrastructure as it is done now.

So really all this patent reveals is how RIM has optimized their new WebKit browser. It's not a new technology on the BlackBerry platform. This fits with comments I've read from Balsillie stating the BlackBerry WebKit browser is amazingly fast by WebKit standards due to the use of RIM's infrastructure.

It's IPPP with three P's and I don't think you're quite right where you say "RIM then uses their infrastructure to translate your optimized IPP [sic] request into a proper HTTP request that public web servers understand." One reason is that an app can make a HTTPS request over IPPP and set the connection option to force end-to-end encryption from the device right through to the target web server. In this case the BIS-B (or BES/MDS) has no idea what the heck your request is doing and certainly can't be doing any sort of translation. IPPP is better thought of as a compressing tunnel from the device all the way through to the BIS-B (or BES/MDS).

You're definitely right on IPPP spelling, my mistake.

Certainly for an HTTPS connection the process I described would be impossible. However that's not to say that RIM uses the same process for handling HTTP and HTTPS connections. If IPPP is just a compressing tunnel, why can I only send HTTP(S) data through it? From your description it would sit somewhere below layer 5 in the simplified 5 layer OSI model, probably between 2 and 3. RIM would probably recommend running any TCP connection through it, maybe even any IP connection.

I'm not an expert on the subject, it's possible you're right, but based on my reading of various developer docs it's a little more than just a compressed encrypted tunnel.

An application can establish a generic socket connection (i.e. not HTTP or HTTPS) that runs over IPPP as far as the BIS-B or BES/MDS and from there to the ultimate endpoint. This fits with my description of IPPP as a compressing (and in the case of BES/MDS, encrypting) tunnel technology.

I was embarassed watching the side by side comparison with the droid on youtube. ohhhh the shame and pain I felt. Wish I could take those 5 minutes back that I lost in my life,...

Is there really a ton of prior art for a proxy server being set on the handheld device itself? That appears to be what's novel here.

Well this sounds like a great "ideal". I hope RIM get's it developed and in the wild soon. I hate having to wait a minute or longer to load any web page I try going to.

Any speed boost will help RIM!! They should allow this to be implemented on apps too so apps can pull in information faster

Correct me if i'm wrong, but doesn't this patent imply that the BB will receive the incoming data to render the webpage, except that it will go to the built-in proxy server on the device first, NOT the actual browser...then the proxy will send that data to a gateway on RIM's BIS network...the gateway will compress and reformat and then send it back to the device's proxy, which will then send to the actual browser itself.

If this is right, then it's adding another layer of transport before the page appears on your screen. Which is fine, as long as the time it takes to process at the gateway and come back is quicker than the device rendering the page itself. But I can't help but think that is redundant in some ways, but what do I know...I'm not an expert.

This also brings up the possibility that if BIS/BES is down, you're not getting web access regardless of the health of your actual mobile network.

For those that do know something about is this better than what Opera or Bolt is doing? As I understand it, those browsers do the requests, compression, and delivery all on the servers before it comes to your phone. RIM's solution implies that the phone itself is doing some work? But doesn't that defeat the purpose of all this?

I'm confused.

This patent talks about working in tandem with a RIM proxy server, or exactly like Bolt and Opera Mini work. Just like Bolt and Opera Mini, a server outage would cause the browser to stop working. This is also how the current BB browser works.

well it sounds like the newest phones that come out in the future will benefit from any changes now, cuz if it takes battery life it would most likely have a newer type of battery and it will come with the latest hardware already to support any changes they make so I'm just waiting to see wat RIM comes out with in the near future but its funny cuz all you hear is sum bitchez cryin that theyre out of the closet and goin Android HaHaHa, so go with ur Androids and get the hell outta here!!! HaHaHa!

I'm for real seek about all the crap about rim and the devices working hard trying and trying doing this or that and at the end frustation because still a lot of bugs or fix in one area and the rest worse. I do own 2 bb storm 2 and tha was the worse discition i ever made in my live.well live and learn.right now i do own the iphone 3gs and i tell you this there is a lot of frustation out there to get one more and in top of that pay for it screw verizon and the black berry, rigth now i'm very happy all the apps work perfect most of them free, internet browser fantastic no more frustation

I will be thrilled when our browser speed increases..there's still days when im loath my phones speed but overall I LOVE THIS PHONE

My first reaction; "yes!"

My second reaction; these are the same clever people who tried to reinvent touch screens with "click thru" technology.

Just do what other smart phone people are doing RIM! Why is it so hard for you to figure out a damn browser?

Its about time they try to speed up the browsing experience. It will probably move up from horrible to mediocre. Even a low spec android phone puts the top BB phone to shame. If the browser wasnt so horrible I probably wouldnt switch phones in my up coming upgrade.

So I take it this will likely require new hardware to even work? Doesn't seem like current handsets would be able to take advantage of this - unless it's all software based???