Reply to thread

Message

G'Day Aaron, thanks for weighing in. Don't let the Spanish flag fool you I'm an Aussie in spain .

First up I think you might be mis-informed on one thing: I see in your post that you have repeated the idea that all this is is a way to auto submit the logs: "The important lesson here is that it's not clear cut that an automatic error log submission system would benefit the project. And it is my opinion that it would not.".

To repeat, I am not talking about an primitive auto log and/or comment submission system. Please see the referenced codeproject and take up the suggestion to run a test on it. If you have never used auto submission of debug info systems before this will amaze you, as it did me the first time I used it.

Some hypothetical questions here I am not actually looking for you to answer them:

1) As a developer and given the choice between debugging a crash/hang problem of some software with your development environments debugger, or alternatively going without it and only allowed to use logging output/event log to track down the offending source code - which of the two would you choose? I think most programmers would take the debugger.

2) If the logging output does not contain the information needed to debug, what do you do.. do you turn to the debugger?

All auto submitted crash reporting does is give the developer the power to debug the crash that the user exprienced in their development environment, as, where and how it happened on the end users machine.. .looking at where it happened in the the source code along with all variables and stack trace availabel to click through and inspect. Of course you can do more, but that is its greatest core strength of what I am talking about.

See my earlier in this thread comments on scripts and submission. This information is also helpful satisitically and can help improve the system, track down the cough, less adept developers. Auso submisison does not take more developer time, it saves it. Again we are not simply talking about sending logs and user comments and some system state information. We are talking debugging info for your development environment.

See my earlier in this thread comments on high profile open source projects that have turned to these submission systems for exactly the opposite reason. Even better chat to their diverse group of developers I am sure they would share their experiences and inform as to why they went to using them.

Cheers,
Keith

Just a quick follow-up: There is some real or percieved developer fear here that auto submission leads to high volume of data. Note You can only usually submit a report when your system crashes or hangs. All other reports, user interface requests etc etc cannot use the auto system (unless they cause crashes). Your not going to get massive high volume unless the software crashes that frequently. Might be the case in the first week the submission system goes live, and after that a new crash free release comes out...

Think the little dialog "send to microsoft box" that pops up when MP or other programs crash. Only the debgging info dumps needed by a development environment are sent to MP developers, not microsoft, where they can actually do some good.

<blockquote data-quote="krntea" data-source="post: 161490" data-attributes="member: 28550">G'Day Aaron, thanks for weighing in. Don't let the Spanish flag fool you I'm an Aussie in spain <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite1" alt=":)" title="Smile&nbsp; &nbsp; :)" loading="lazy" data-shortname=":)" />.First up I think you might be mis-informed on one thing: I see in your post that you have repeated the idea that all this is is a way to auto submit the logs: &quot;The important lesson here is that it's not clear cut that an automatic error log submission system would benefit the project.&nbsp; And it is my opinion that it would not.&quot;. To repeat, I am not talking about an primitive auto log and/or comment submission system. Please see the referenced codeproject and take up the suggestion to run a test on it. If you have never used auto submission of debug info systems before this will amaze you, as it did me the first time I used it.Some hypothetical questions here I am not actually looking for you to answer them: 1) As a developer and given the choice between debugging a crash/hang problem of some software with your development environments debugger, or alternatively going without it and only allowed to use logging output/event log to track down the offending source code&nbsp; - which of the two would you choose?&nbsp; I think most programmers would take the debugger.2) If the logging output does not contain the information needed to debug, what do you do.. do you turn to the debugger?&nbsp; All auto submitted crash reporting does is give the developer the power to debug the crash that the user exprienced in their development environment, as, where and how it happened on the end users machine.. .looking at where it happened in the the source code along with all variables and stack trace availabel to click through and inspect. Of course you can do more, but that is its greatest core strength of what I am talking about.See my earlier in this thread comments on scripts and submission. This information is also helpful satisitically and can help improve the system, track down the cough, less adept developers. Auso submisison does not take more developer time, it saves it. Again we are not simply talking about sending logs and user comments and some system state information. We are talking debugging info for your development environment. See my earlier in this thread comments on high profile open source projects that have turned to these submission systems for exactly the opposite reason. Even better chat to their diverse group of developers I am sure they would share their experiences and inform as to why they went to using them.Cheers,KeithJust a quick follow-up: There is some real or percieved developer fear here that auto submission leads to high volume of data. Note You can only usually submit a report when your system crashes or hangs.&nbsp; All other reports, user interface requests etc etc cannot use the auto system (unless they cause crashes).&nbsp; Your not going to get massive high volume unless the software crashes that frequently. Might be the case in the first week the submission system goes live, and after that a new crash free release comes out... Think the little dialog &quot;send to microsoft box&quot; that pops up when MP or other programs crash. Only the debgging info dumps needed by a development environment are sent to MP developers, not microsoft, where they can actually do some good.</blockquote>

[QUOTE="krntea, post: 161490, member: 28550"] G'Day Aaron, thanks for weighing in. Don't let the Spanish flag fool you I'm an Aussie in spain :-). First up I think you might be mis-informed on one thing: I see in your post that you have repeated the idea that all this is is a way to auto submit the logs: "The important lesson here is that it's not clear cut that an automatic error log submission system would benefit the project. And it is my opinion that it would not.". To repeat, I am [B]not[/B] talking about an primitive auto log and/or comment submission system. Please see the referenced codeproject and take up the suggestion to run a test on it. If you have never used auto submission of debug info systems before this will amaze you, as it did me the first time I used it. Some hypothetical questions here I am not actually looking for you to answer them: 1) As a developer and given the choice between debugging a crash/hang problem of some software with your development environments debugger, or alternatively going without it and only allowed to use logging output/event log to track down the offending source code - which of the two would you choose? I think most programmers would take the debugger. 2) If the logging output does not contain the information needed to debug, what do you do.. do you turn to the debugger? All auto submitted crash reporting does is give the developer the power to debug the crash that the user exprienced [B]in their development environment[/B], as, where and how it happened on the end users machine.. .looking at where it happened in the the source code along with all variables and stack trace availabel to click through and inspect. Of course you can do more, but that is its greatest core strength of what I am talking about. See my earlier in this thread comments on scripts and submission. This information is also helpful satisitically and can help improve the system, track down the cough, less adept developers. Auso submisison does not take more developer time, it saves it. Again we are not simply talking about sending logs and user comments and some system state information. We are talking debugging info for your development environment. See my earlier in this thread comments on high profile open source projects that have turned to these submission systems for exactly the opposite reason. Even better chat to their diverse group of developers I am sure they would share their experiences and inform as to why they went to using them. Cheers, Keith Just a quick follow-up: There is some real or percieved developer fear here that auto submission leads to high volume of data. Note You can [b]only[/b] usually submit a report [b]when your system crashes or hangs[/b]. All other reports, user interface requests etc etc cannot use the auto system (unless they cause crashes). Your not going to get massive high volume unless the software crashes that frequently. Might be the case in the first week the submission system goes live, and after that a new crash free release comes out... Think the little dialog "send to microsoft box" that pops up when MP or other programs crash. Only the debgging info dumps needed by a development environment are sent to MP developers, not microsoft, where they can actually do some good. [/QUOTE]