Re: Re: Suggest Bounty for Defects

"Ken Kahn" <[email protected]> Sun, 2 Jan 2005 22:09:57 -0000
Newsgroups gmane.education.weblabs.requests
Message-ID <000a01c4f117$ccdace10$57f34ed5@ESTLAP05>
------=_NextPart_000_0007_01C4F117.CBC1BCF0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi all.

I think the Logotron testers are doing the sensible thing -- searching for =
bugs that are the most likely to be encountered by customers. Most customer=
s whether at home or in schools don't try the things that you and Ylva are =
doing. Now Logotron is wearing two hats here: publisher of ToonTalk in vari=
ous parts of the world and WebLabs partner. They also did a good amount of =
WebLabs/ToonTalk specific testing.

I agree that it is very frustrating when ToonTalk fails -- especially when =
it is core functionality that breaks. But I don't believe the bugs that you=
 and I encounter reflect normal use. You had one where a bird who had a cop=
y of her nest in a box that was dropped on an erased number got confused. I=
 had one recently where I trained a robot to break up a box and then vacuum=
ed up the half that ToonTalk thought was "My Box".

This may be another argument against a substantial bounty because people wi=
ll look for bugs in "obscure places" and divert resources from making the m=
ore common usage patterns more robust and reliable.

Sorry to hear that ToonTalk bugs are preventing you from continuing your pr=
oject. I'll fix any reported bug but I understand that the appearance and r=
eporting of bugs can slow one down too much.

Best,

-ken
  ----- Original Message -----=20
  From: Jack Waugh=20
  To: [email protected]=20
  Sent: Saturday, January 01, 2005 8:50 PM
  Subject: [WebLabs-requests] Re: Suggest Bounty for Defects



  --- In http://groups.yahoo.com/group/WebLabs-requests/message/797,
  "Ken Kahn" <kenkahn@t...> wrote:

  > Regarding Ylva's bugs one of them
  . . . > only manifested itself if all the following=20
  > happened:
  [several conditions]> . It took=20
  > me quite a while to reproduce the
  > problem because it was a combination=20
  > of many circumstances.

  The faults I run into don't require real-time circumstances; they
  reflect straightforward defects in recording or interpreting the programs=
.

  > The other bug manifested itself only if you saved=20
  > a notebook with user pictures in it that
  > were saved after being created=20
  > but before they had ever been displayed.
  > (Pretty hard to do.)

  None of the faults I run into involve pictures.  I'm not doing
  anything with pictures, nor remote controls, nor sensors.  Just
  straightforward robots, trucks, bombs, birds, nests, boxes, text
  (treated atomically), and numbers.  Basic stuff that should have been
  working before the code went into beta test.

  > That makes it very hard for professional testers to catch bugs.

  It's very easy for researchers noodling around with styles in which to
  use birds and nests to be stopped in their work due to overwhelming
  numbers of severe defects.

  > Logotron=20
  > now has 2 people putting in a large
  > amount of time testing ToonTalk and=20
  > they've been having a hard time finding
  > bugs in last the month or so.

  Then their wages are going to waste.  If the idea is to document the
  defects, Logotron had best fire those testers from the ToonTalk
  project and redirect their wages into the jackpot system I suggested;
  this would produce many more defect reports per Euro spent.

  > But I find them regularly as does anyone doing a large size project.

  I'm doing a very small project, and I can't get anywhere because
  robots don't behave the way they're supposed to often enough to use
  them to test the simplest ideas.

  > Regarding the bounty idea, Animated
  > Programs can't afford it but I'll=20
  > speak to folks at Logotron about this on my next visit.

  Why wait until then?

  > But I wonder how=20
  > many people there are that encounter bugs but don't report them but=20
  > would if there was such an incentive.

  I would if I thought I could earn a living at it.  If the bounty were
  high enough, I could, but I'm not sure how high is high enough in that
  sense.  I should think the wages of the current two people who are
  putting in large amounts of time would be enough.  I expect I'd earn
  at least half the bounty.  My experience with various software
  packages (earlier Microsoft Word, Rational Rose, for two examples) is
  that even though other people don't complain, if I so much as touch
  the software, it breaks.

  > And how many people stop using=20
  > ToonTalk due to bugs but with the right
  > reward system would continue.

  The only thing that could make me continue using it without
  compensation would be a form of mental illness.






---------------------------------------------------------------------------=
---
  Yahoo! Groups Links

    a.. To visit your group on the web, go to:
    http://groups.yahoo.com/group/WebLabs-requests/
=20=20=20=20=20=20
    b.. To unsubscribe from this group, send an email to:
    [email protected]
=20=20=20=20=20=20
    c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service=
.=20


------=_NextPart_000_0007_01C4F117.CBC1BCF0
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2900.2523" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>



<DIV><FONT face="Comic Sans MS" size=2>Hi all.</FONT></DIV>
<DIV><FONT face="Comic Sans MS" size=2></FONT>&nbsp;</DIV>
<DIV><FONT face="Comic Sans MS" size=2>I think the Logotron testers are doing 
the sensible thing -- searching for bugs that are the most likely to be 
encountered by customers. Most customers whether at home or in schools don't try 
the things that you and Ylva are doing. Now Logotron is wearing two hats here: 
publisher of ToonTalk in various parts of the world and WebLabs partner. They 
also did a good amount of WebLabs/ToonTalk specific testing.</FONT></DIV>
<DIV><FONT face="Comic Sans MS" size=2></FONT>&nbsp;</DIV>
<DIV><FONT face="Comic Sans MS" size=2>I agree that it is very frustrating when 
ToonTalk fails -- especially when it is core functionality that breaks. But I 
don't believe the bugs that you and I encounter reflect normal use. You had one 
where a bird who had a copy of her nest in a box that was dropped on an erased 
number got confused. I had one recently where I trained a robot to break up a 
box and then vacuumed up the half that ToonTalk thought was "My 
Box".</FONT></DIV>
<DIV><FONT face="Comic Sans MS" size=2></FONT>&nbsp;</DIV>
<DIV><FONT face="Comic Sans MS" size=2>This may be another argument against a 
substantial bounty because people will look for bugs in "obscure places" and 
divert resources from making the more common usage patterns more robust and 
reliable.</FONT></DIV>
<DIV><FONT face="Comic Sans MS" size=2></FONT>&nbsp;</DIV>
<DIV><FONT face="Comic Sans MS" size=2>Sorry to hear that ToonTalk bugs are 
preventing you from continuing your project. I'll fix any reported bug but I 
understand that the appearance and reporting of bugs can slow one down too 
much.</FONT></DIV>
<DIV><FONT face="Comic Sans MS" size=2></FONT>&nbsp;</DIV>
<DIV><FONT face="Comic Sans MS" size=2>Best,</FONT></DIV>
<DIV><FONT face="Comic Sans MS" size=2></FONT>&nbsp;</DIV>
<DIV><FONT face="Comic Sans MS" size=2>-ken</FONT></DIV>
<BLOCKQUOTE 
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
  <DIV style="FONT: 10pt arial">----- Original Message ----- </DIV>
  <DIV 
  style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B> 
  <A [email protected] 
  href="mailto:[email protected]">Jack Waugh</A> </DIV>
  <DIV style="FONT: 10pt arial"><B>To:</B> <A 
  [email protected] 
  href="mailto:[email protected]">[email protected]</A> 
  </DIV>
  <DIV style="FONT: 10pt arial"><B>Sent:</B> Saturday, January 01, 2005 8:50 
  PM</DIV>
  <DIV style="FONT: 10pt arial"><B>Subject:</B> [WebLabs-requests] Re: Suggest 
  Bounty for Defects</DIV>
  <DIV><BR></DIV><TT><BR>--- In <A 
  href="http://groups.yahoo.com/group/WebLabs-requests/message/797,">http://groups.yahoo.com/group/WebLabs-requests/message/797,</A><BR>"Ken 
  Kahn" &lt;<A href="mailto:kenkahn@t">kenkahn@t</A>...&gt; wrote:<BR><BR>&gt; 
  Regarding Ylva's bugs one of them<BR>. . . &gt; only manifested itself if all 
  the following <BR>&gt; happened:<BR>[several conditions]&gt; . It took 
  <BR>&gt; me quite a while to reproduce the<BR>&gt; problem because it was a 
  combination <BR>&gt; of many circumstances.<BR><BR>The faults I run into don't 
  require real-time circumstances; they<BR>reflect straightforward defects in 
  recording or interpreting the programs.<BR><BR>&gt; The other bug manifested 
  itself only if you saved <BR>&gt; a notebook with user pictures in it 
  that<BR>&gt; were saved after being created <BR>&gt; but before they had ever 
  been displayed.<BR>&gt; (Pretty hard to do.)<BR><BR>None of the faults I run 
  into involve pictures.&nbsp; I'm not doing<BR>anything with pictures, nor 
  remote controls, nor sensors.&nbsp; Just<BR>straightforward robots, trucks, 
  bombs, birds, nests, boxes, text<BR>(treated atomically), and numbers.&nbsp; 
  Basic stuff that should have been<BR>working before the code went into beta 
  test.<BR><BR>&gt; That makes it very hard for professional testers to catch 
  bugs.<BR><BR>It's very easy for researchers noodling around with styles in 
  which to<BR>use birds and nests to be stopped in their work due to 
  overwhelming<BR>numbers of severe defects.<BR><BR>&gt; Logotron <BR>&gt; now 
  has 2 people putting in a large<BR>&gt; amount of time testing ToonTalk and 
  <BR>&gt; they've been having a hard time finding<BR>&gt; bugs in last the 
  month or so.<BR><BR>Then their wages are going to waste.&nbsp; If the idea is 
  to document the<BR>defects, Logotron had best fire those testers from the 
  ToonTalk<BR>project and redirect their wages into the jackpot system I 
  suggested;<BR>this would produce many more defect reports per Euro 
  spent.<BR><BR>&gt; But I find them regularly as does anyone doing a large size 
  project.<BR><BR>I'm doing a very small project, and I can't get anywhere 
  because<BR>robots don't behave the way they're supposed to often enough to 
  use<BR>them to test the simplest ideas.<BR><BR>&gt; Regarding the bounty idea, 
  Animated<BR>&gt; Programs can't afford it but I'll <BR>&gt; speak to folks at 
  Logotron about this on my next visit.<BR><BR>Why wait until then?<BR><BR>&gt; 
  But I wonder how <BR>&gt; many people there are that encounter bugs but don't 
  report them but <BR>&gt; would if there was such an incentive.<BR><BR>I would 
  if I thought I could earn a living at it.&nbsp; If the bounty were<BR>high 
  enough, I could, but I'm not sure how high is high enough in 
  that<BR>sense.&nbsp; I should think the wages of the current two people who 
  are<BR>putting in large amounts of time would be enough.&nbsp; I expect I'd 
  earn<BR>at least half the bounty.&nbsp; My experience with various 
  software<BR>packages (earlier Microsoft Word, Rational Rose, for two examples) 
  is<BR>that even though other people don't complain, if I so much as 
  touch<BR>the software, it breaks.<BR><BR>&gt; And how many people stop using 
  <BR>&gt; ToonTalk due to bugs but with the right<BR>&gt; reward system would 
  continue.<BR><BR>The only thing that could make me continue using it 
  without<BR>compensation would be a form of mental 
  illness.<BR><BR><BR><BR><BR></TT>





</BODY></HTML>

------=_NextPart_000_0007_01C4F117.CBC1BCF0--