Rent a Signal service development and suggestions - page 20

 

^ Thanks. The developers need to sit down together, and audit all the pairs for pips calculation. This should be the core and the very fundamental basic of the signal providing service. This is not the first time I've mentioned and pointed out this problem. Before it was on the JPY pairs.

Get this nailed down first, and I think a lot of your subsequent problems will be solved. And it will also instill confidence in both the sellers and buyers to continue and provide more feedback. Right now, I think a lot of sellers are seeing wrong results, and lose motivation to continue.

Keep up the good work.

 

JB... FOR THE LAST TIME THIS IS beta!

There are no buyers...

You must understand this is not a presentation to sellers for them to evaluate with...you are a BETA TESTER!

By the way...you and Wayne, are the #1 traders eventually

ES

Kenny Rogers:
^ Thanks. The developers need to sit down together, and audit all the pairs for pips calculation. This should be the core and the very fundamental basic of the signal providing service. This is not the first time I've mentioned and pointed out this problem. Before it was on the JPY pairs.

Get this nailed down first, and I think a lot of your subsequent problems will be solved. And it will also instill confidence in both the sellers and buyers to continue and provide more feedback. Right now, I think a lot of sellers are seeing wrong results, and lose motivation to continue.

Keep up the good work.
 

zupcon,

I have pondered this post of yours, I quoted below...and I have always had great respect for your knoweledge.

I can agree that MT4 and its structure is well...questionable. I understand your points (the scenarios in plain speak) and I share my well...disappointment with MT4. Len...will not answer or reply to my public questions in the past, so I have given up..Perhaps I am not in his league when it comes to programming and such...I am just a user that has concerns about the dealer only benefits and other structural and marekting plug-in ploys. The very partial C++ programming language used called mql...from what I understand misses a lot and leaves some vulnerability.

zupcon...if you had to solve these problems you point out and to prevent this dishonestly...how would you stablize the situation? No you must keep MT4 too! This global unregulated market without a standardized feed is troubling to say the least.

ES

zupcon:
There's no danger of sellers EA's being decompiled, the only "danger" to the seller is that someone backward engineers aspects of the system, but that can happen with any sort of signals service.

If I where an unscrululous seller this is what I would do. I would decompile the EA that sends the signals, and Id modify it to knock a couple of pips off the true buy price, and add a couple of pips to the true sell price, giving me a slightly better entry than I in fact obtained. I'd screw a couple of pips on the exit as well . I'd also modify the time stamp on the order. Id also ensure any slippage was generally in my favour . Therefore the results being recorded and audited would not be a true reflection of my actual trades. Its surprising how many break even type systems perform so much better when you eliminate the spread LOL

When new digitals system gets confirmation of the trade some time later due to latency, the system has no idea if that signal was delayed due to real latency, or if the delay is due to tampering.

The problem with these sorts of systems is this, does the system simply "believe" the trade details being sent by the sellers EA (and if so, lend itself open to potential abuse) or does the system check every incoming signal against a data feed (which of course might differ slightly from the sellers data feed, and be subject to latency etc causing sellers to constantly question and complain entry and exit prices, and slippage etc)

These systems are certainly more suceptible to abuse if used for short term trades.

Of course, decompiling the EA would also provide details of the server and scrips being used to accept incoming trades, and whilst shoulnt be a concern if the system is secure, its a vunrability which might be exploited

Cllearly these types of frauds cant run indefinately, a system seller might be showing good gains, but if the subscribers cant achieve the same fills then they'll soon unsubscribe, however it could have the potential to undermine the credability of such a system.

If the system is geared towards longer term trades where a few pips here or there dont make much difference its not as much of an issue, but it still gives the unscrupulous seller a slight edge

Of course any competitor could save themselves some work by decompiling the EA's.

I just cant understand why any commercial project would use a product that could be decompiled with such ease
 
ElectricSavant:
zupcon,

zupcon...if you had to solve these problems you point out and to prevent this dishonestly...how would you stablize the situation? No you must keep MT4 too!

ES

I'm not involved in the beta test, so I'll have to make some assumptions. I assume the software is comprised as an MQL executable and a dll file, and I'm assuming the dll is responsible for transmission of the signal passed to it by the MQL file.

The contents of the dll should be secure, the MQL executable is open to tampering. So a few things off the top of my head:

The dll could check the properties of the MQL executable file such as file size etc, if the file's been tapmered with, its size will probably change. You could get the dll file to read from a particular byte position within the executable to checkthat nothing had been inserted or deleted. Fairly basic stuff, easy to impliment and would catch out the script kiddies

There are probably a few tricks you could play with the MQL file, such as forcing the user to pass redundant arguments to functions in the dll. For example, you could force the MQL file to pass in the date and time of the trade, and the dll could compare this against the system clock. Any attempt to backdate a trade could be identified. Thats just a simple example of the type of thing you might do.

Encrypting all trafic between the dll and server is probably sensible, that would certainly complicate attempts to spoof the functionality of the dll.

Unfortunately the only way to catch potential problems is rigorous auditing of the system, this might include checking all trades against an independant data feed, identifying trades who always seam to benefit from slippage on both entry and exit, identifying trades where latency between the time the trade is taken on the suppliers terminal, and the time the signal is recieved by TSD is greater than normal. Some form of basic sanity checking is always useful, for example, is it likely to achieve a 470 pip profit from a trade only open for 20 minutes etc.

I'm unclear as to whose responsible for processing payments in this system, and what payment methods are available ? policies on chargbacks etc. I've seen some seriously creative frauds built around weaknesses in payment processing

There are probably quite a few people out there who dont want this to work, such IB's, signal sellers, EA sellers, the ledgendary el cid (LOL), not to mention competitors, other forums, and possibly even brokers.

Should be interesting to see how this develops

 

I will tell you what folks...

If this security is beyond the realm of the developers and if zupcon is taking on any clients and if I were rent a signal or tsd...I WOULD HIRE HIM IN A HEARTBEAT.

But...

I am not suggesting that the present development is inferior or anything like that...but in my opinion this is a million dollar idea and the caliber of professionals like zupcon may in fact be interested in this project...now or at a later time...we must hope for that..

ES

P.S. Or maybe I am just dazzled at stuff I do not understand! ☺

 

I'm fully aware this is in beta testing and there are no buyers.

But as a beta tester, I'm evaluating the whole system as a buyer and seller, and giving my feedback to ND. This is what a beta tester does. They don't need a cheerleader.

But if someone consistently brings up the same type of problems, I think it is beneficial to do a thorough job of cleaning up the problem across the board. Or it will start to turn off other beta testers from providing feedback...which is probably the case at the moment. When sellers trade, and then see totally different results on the RAS website, they don't take the time or effort to look in detail why it is happening. I had to look into OTHER sellers statements and identify the problem.

Being a beta tester is my contribution to this forum and ND, I don't have plans to be a seller. It would take a lot more forward testing before I would even think about providing signals.

 

I could never be a successful cheerleader as I am not cute enough and I agree with all you say JB...

But I think they need a few more weeks to get to all of this. The priorities may not be very clear as explained by ND...they seem to be preoccupied in other items they feel are more important...what that is....who knows?...because this statistical reporting on the Website is very very important...

I think the IBFX..the 5 decimals....the standard and mini's....they are aware of...There seems to be two main projects the "broadcast" and the "website"...

ES

P.S. Loading up the rent a signal with a bunch of testers is actually a very good idea as obviously everything was not conceptualized and thought of...much is being discovered and many changes being made...it truly is a work in progress...You know I am not very graceful at all...Where are my pom pom's? ☺ This stupid bra keeps falling off...and my boxer-shorts are all in a bunch:)

Kenny Rogers:
I'm fully aware this is in beta testing and there are no buyers.

But as a beta tester, I'm evaluating the whole system as a buyer and seller, and giving my feedback to ND. This is what a beta tester does. They don't need a cheerleader.

But if someone consistently brings up the same type of problems, I think it is beneficial to do a thorough job of cleaning up the problem across the board. Or it will start to turn off other beta testers from providing feedback...which is probably the case at the moment. When sellers trade, and then see totally different results on the RAS website, they don't take the time or effort to look in detail why it is happening. I had to look into OTHER sellers statements and identify the problem.

Being a beta tester is my contribution to this forum and ND, I don't have plans to be a seller. It would take a lot more forward testing before I would even think about providing signals.
With new updated version of RAS tool: this comparing process is continuing. How many updates? Impossible to count even. It was 2 updates of RAS tool for the last 4 days and 2 updates of server parts of this tool (for the last 4 days as well). All updates were done based on fixing the errors reported. RAS tool is updated because: members reported about the error, I confirm this error with the other testers and sending all the information to the developers and we receive new version with fixed error. It is beta testing.
 

GIGO. Better get the calculation right first. Or the statistics mean nothing.

That's all I'm try to say.

 
ElectricSavant:
JB... FOR THE LAST TIME THIS IS beta!

There are no buyers...

You must understand this is not a presentation to sellers for them to evaluate with...you are a BETA TESTER!

By the way...you and Wayne, are the #1 traders eventually ☺

ES

LOL, Not any more, I'm back to last again.

I think Alpari-UK accounts are giving the developers a headache?

Cheers

Wayne

 
Kenny Rogers:
GIGO. Better get the calculation right first. Or the statistics mean nothing. That's all I'm try to say.

I would have to concur here? It's like having a vehicle and the wrong fuel for it?

No point in having the vehicle then, is there?

So, lets get the fuel right and then decide where we want to go.

Stats and the calculation thereof should take a decent coder less than a day to complete and have functioning 100%, then the Beta Testers like us can stop worrying about why our results are not the same as those published and get on with helping to resolve the real issues.

Reason: