Fog Creek Software
Discussion Board




Working remotely without pain ?

Hello,

There are many applications for remote connections. Did any of you succeed to do it without noticing the difference ? I don't mean *any* difference (say, diff in colors and resolution aren't important), what I actually care is diff in speed. So far, a couple of programs I tried all gave a noticeable and painful delay for all my actions ...

What about the built-in service of Windows XP ?

What I'm trying to achieve is to work (and not mimic working) remotely from the laptop sitting inside the LAN. The network connection is wireless. Is it possible or there's no chance ?

Thank you !

Evgeny Goldin
Sunday, December 01, 2002

Wireless netwrk connections are slow. The old standard was 1Mbps and that was never achieved. To make things worse the speed drops drastically as you move away from the base station.

My advice is to get a long run of CAT5 , two RJ45 connectors, and borrow a crimper.

Stephen Jones
Sunday, December 01, 2002

Current wireless products are much better, though it always depends on the environment.  You should get in a reasonable environment the quoted 11Mb using 802.11b.

Simon Lucy
Sunday, December 01, 2002

The tests I've seen for 802.11b suggest an average much, much slower than the quoted max.  The same roonm as the base station might get you the max speed. As you said, it all depends.

Stephen Jones
Sunday, December 01, 2002

I get a full 11Mbps with my D-Link wireless AP as long as I'm within 50 feet or so. In other words, it pretty much only saves me the hassle of a cable running across my floor. It goes about 200-300 feet before complete signal loss. But as long as a signal is there - no matter how weak - the Remote Desktop Services (aka, Terminal Services) of Windows XP Pro work just fine, assuming the settings are tweeked correctly. I use RDS over a VPN connection with a cable modem to work from home and, other than being limited to a single monitor with RDS, there's no difference from being in the office.

Having said that, I actually don't use RDS to do anything other than administrative type work. All coding I do running locally so that I get back my second monitor. The only issue there was using a revision-control system built to work over TCP/IP (that is, getting *off* of VSS). Sure, the initial "Get Latest" isn't as zippy as being in the office, but it's still more than adequate. And once that's complete, the remote connection is no longer needed until check-in time.

Ryan LaNeve
Sunday, December 01, 2002

I don't wear a pyramid hat made of tin foil, believe in conspiracy theories, and all that - but on the subject of wireless I've got to admit that I'm a little paranoid.

It started out when a former co-worker swore he could "feel" the wireless signal in his home network.  It freaked him out, so he reverted back to a hardwired network. I looked into the subject, and a study commissioned by the Canadian government found that a small percentage of people could feel the signals.

Another study (published in Nature, I believe) found that the use cell phones will produce a temperature elevation in the brain.  The temperature elevation is slight in adults, more pronounced in teenagers, and significant in young children. The long term effects are not known.

Since I have young children, I decided against a wireless home network. I'm not trying to spread my paranoia, but I'm not sure that the march toward onmnipresent wireless signals a good thing in the long run.

Nick Hebb
Sunday, December 01, 2002

The reason that cell phones can elevate skin temperature is that it is so close to the skin.  Since radio waves dissipate as 1/r^2, where r is the distance between the transmitter (phone) and receiver (skin), the proximity of an 802.11 wireless transmitter (30 mW) to your face is far less than that of a cell phone (750-1000 mW).

Bottom line:  I'd doubt anyone's claim that they can feel 802.11 E&M waves when they are in the vicinity.

In fact, "the vicinity" is much larger than you might think.  I went "war driving" several times in the last month, just to see what I could see.  Put the laptop in the car, put my Cisco card in "listen mode"  and fired up "AirSnort" http://airsnort.shmoo.com/ .

I found that while driving my kids to football practice I encoutered on average 3 wireless access points per mile along the drive.  There might be more, but I was driving at 30-40 mph, so I may have missed some along the way.  About 1/3 of those were WEP encrypted, the others were in the clear.

So, in other words, as you drive along the streets in your cars, you're encountering other people's 802.11 wireless stations.  Does your friend feel those?

Additionally, I was trying to crack my own WEP at home, just to see how vulnerable the known "initialization vector" vulnerability is.  I ran many FTP downloads to get the packet capture size up, but never found any "interesting" packets (see the airsnort page for the definition of "interesting").  It turns out that thought perhaps a bad design, the 802.11 vulnerability is not so bad as reported - at least when used as a home network.

Nat Ersoz
Sunday, December 01, 2002

There has been a number of threads here.

As mentioned, by far and away the best solution is windows terminal server. That is the RDP protocal. It is also the same stuff used for win xp remote support.

I use it daily, and it rocks. As mentioned in this thread, you can hardly tell the difference. In other words, anything faster that a 56k modem is great. TS is even workable on a modem.

Since the wireless network is quite a bit faster, then 56k, then RDP is ideal….

If your server is running IIS, then you can actually web-enable any software you have with a few mouse clicks using this technology! (why bother re-writng for the web!).

Albert D. Kallal
Edmonton, Alberta Canada
Kallal@msn.com

Albert D. Kallal
Sunday, December 01, 2002

Sorry, but I have to chime-in on something...

Running a desktop application via Terminal Services/RDP does not, in any way, "web-enable" the application. In my opinion, to say an application is "web-enabled" because it can be run over TS/RDP is a gross misrepresentation and belittles the work of those who do actual conversions from desktop to browser-based application.

I've sat in far too many product demos where the sales rep (or even technical rep in some cases) will utter the phrase, "it runs through a browser", when in fact they're doing nothing more than initiating the TS/RDP connection using the web-interface for TS/RDP. I now walk out of such meetings when such a rediculously misleading statement is made. How can I possibly believe anything else the person says from here on out?

Don't get me wrong - I love TS/RDP, and I certainly use it to run traditional desktop apps from a remote location. But never in a million years would I say "I 'web-enabled' the app".

Ryan LaNeve
Sunday, December 01, 2002

Thank you, guys !

Evgeny Goldin
Monday, December 02, 2002

Actually, I really don’t see why using TS to web enable a product is somehow fake?

Why not use a few mouse clicks to web enable any windows program?

I don’t see this as a misleading concept in any way, shape, or form?

Why is this misleading? By who’s definition is this bad thing?

Certainly, one is not claiming that to run an application across the web, that MUST use HTML..are you?

Just about every company from Oracle on down the line has figured out that a HTML browser is a crap interface. They ALL NOW download some type of client on the pc. That client may, or may not crank out HTML. Oracle forms and all of the products out there that make nice web interfaces ALL NOW download a intelligent client. That is EXACITY the same thing that TS does. Since everyone is doing this..why is this misleading all of a sudden?

The above approach simply means that developer is free to use their best, and most productive tools to build the GUI. The idea somehow that everything after that should be re-written *AGAIN* for the web is nuts. We should get both windows, and the web for the effort of developing an application.  TS gives developers exactly that, and NO NEW tools are required. This is a incredible valuable concept.

Many companies have a small sales force on the road. They bring  in a consulting company, and for what …5, or 10 people on the road they re-write their customer relations system,  or customer order system at HUGE COST?

Why spend huge money when products like TS instantly web enable the product?

In fact, I would go so far as to say that many IT companies are very dishonest, and don’t tell clients about TS, since then they would not get a contract to re-write the software for the web at great cost. (now who is being the snake oil salesman here? hum?).

No, TS is a great way to web enable products with a few mouse clicks. The fact that web developers can be replaced with a few mouse clicks does reduce their value a lot. However, it is not the fact that developers loose jobs because of TS, it the fact that so many companies are getting soaked since some consultants think that TS is misleading the customer in this regards!

It is a disgusting idea, and at the very best a morally reprehensible consulting tactic to suggest that TS is somehow misleading the customer. TS is a fine way to web enable a product, and it does it without one line of the software having to be changed.

I have to differ on this one on a LOT!!.

The real misleading occurring here is the fact that many IT companies do not offer TS as a workable solution, since NO work will result as of offering the concept. It is too good to be true, and thus some people want others to be insulted at this type of suggestion.

It is not a web based solution for the general masses, but for a lot companies with existing  software, in many cases it is the best choice by a country mile.

Albert D. Kallal
Edmonton, Alberta Canada
Kallal@msn.com

Albert D. Kallal
Saturday, December 07, 2002

*  Recent Topics

*  Fog Creek Home