.NET Reflector now belongs to Redgate

[Gabriel Schenker]

It's not just money that should rule the world... If Lutz has worked on a
tool which was and is freely available he has done something good for the
community. On the other hand he certainly has also used other OSS tools
which were implemented by other guys. Or he has got part of his knowledge
from reading blogs where others have published insight into patterns and
practices for free.

Thus Charlie your statement should rather be:
"I presume everyone who is complaining has done some valuable work for the
community." :-))

On Fri, Aug 22, 2008 at 10:04 PM, Charlie Poole <charlie@...> wrote:

>   I presume everyone who is complaining has donated to Lutz for
> his work on Reflector. ;-)
>
> Charlie
>
> > -----Original Message-----
> > From: altdotnet@... <altdotnet%40yahoogroups.com>
> > [mailto:altdotnet@... <altdotnet%40yahoogroups.com>] On
> Behalf Of Sebastien Lambla
> > Sent: Friday, August 22, 2008 5:50 AM
> > To: altdotnet@... <altdotnet%40yahoogroups.com>
> > Subject: Re: [altdotnet] .NET Reflector now belongs to Redgate
> >
> > Reflector has never been open source, neither has testdriven.net
> >
> > I wish them well and hope they'll provide a commercial
> > version, for example letting you dynamically step through
> > decompiled code. THat would be a real time saver when dealing
> > with unavailable code...
> >
> > --------------------------------------------------
> > From: "Gabriel Schenker" <gnschenker@... <gnschenker%40gmail.com>>
> > Sent: Friday, August 22, 2008 1:28 PM
> > To: <altdotnet@... <altdotnet%40yahoogroups.com>>
> > Subject: [altdotnet] .NET Reflector now belongs to Redgate
> >
> > > after years of development Lutz Roeder obviously abandons it's flag
> > > ship an has given (sold?) it to Redgate.
> > > Although Redgate has annouced to provide a free version also in the
> > > future I fear that in the long run you will have to pay for
> > the full
> > > version of Reflector.
> > > Do we have to say goodbye to another indispensable OSS
> > tool? (see e.g.
> > > what happened to NCover and Testdriven.Net)
> > >
> > >
> > > ------------------------------------
> > >
> > > Yahoo! Groups Links
> > >
> > >
> > >
> > >
> >
> > ------------------------------------
> >
> > Yahoo! Groups Links
> >
> >
> >
> >
>
>  
>

[Charlie Poole]

Well, the whining was pretty much all about the money. Hence the nature of
my comment.
 
I appreciate the philosophy behind your comment though!
 
Charlie
 
PS: Reflector has never been Open Source.
 
  _____  

From: altdotnet@... [mailto:altdotnet@...] On Behalf
Of Gabriel Schenker
Sent: Friday, August 22, 2008 9:05 PM
To: altdotnet@...
Subject: Re: [altdotnet] .NET Reflector now belongs to Redgate



It's not just money that should rule the world... If Lutz has worked on a
tool which was and is freely available he has done something good for the
community. On the other hand he certainly has also used other OSS tools
which were implemented by other guys. Or he has got part of his knowledge
from reading blogs where others have published insight into patterns and
practices for free.

Thus Charlie your statement should rather be:
"I presume everyone who is complaining has done some valuable work for the
community." :-))


On Fri, Aug 22, 2008 at 10:04 PM, Charlie Poole <charlie@...> wrote:


I presume everyone who is complaining has donated to Lutz for
his work on Reflector. ;-)

Charlie



> -----Original Message-----
> From: altdotnet@... <mailto:altdotnet%40yahoogroups.com>  
> [mailto:altdotnet@... <mailto:altdotnet%40yahoogroups.com> ]
On Behalf Of Sebastien Lambla
> Sent: Friday, August 22, 2008 5:50 AM
> To: altdotnet@... <mailto:altdotnet%40yahoogroups.com>

> Subject: Re: [altdotnet] .NET Reflector now belongs to Redgate
>
> Reflector has never been open source, neither has testdriven.net
>
> I wish them well and hope they'll provide a commercial
> version, for example letting you dynamically step through
> decompiled code. THat would be a real time saver when dealing
> with unavailable code...
>
> --------------------------------------------------
> From: "Gabriel Schenker" <gnschenker@...

<mailto:gnschenker%40gmail.com> >

> Sent: Friday, August 22, 2008 1:28 PM
> To: <altdotnet@... <mailto:altdotnet%40yahoogroups.com> >
> Subject: [altdotnet] .NET Reflector now belongs to Redgate
>
> > after years of development Lutz Roeder obviously abandons it's flag
> > ship an has given (sold?) it to Redgate.
> > Although Redgate has annouced to provide a free version also in the
> > future I fear that in the long run you will have to pay for
> the full
> > version of Reflector.
> > Do we have to say goodbye to another indispensable OSS
> tool? (see e.g.
> > what happened to NCover and Testdriven.Net)
> >
> >
> > ------------------------------------
> >
> > Yahoo! Groups Links
> >
> >
> >
> >
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>

 

[Gael Fraiteur]

It's not on my desktop but on my quick launch toolbar ;).

What I meant there is that the software we produce have no dependency
on Reflector. If you released Castle with a similar license, you
would probably have no customer at all.

The tool is essential but fully interchangeable with another tool
fullfilling the same function. If Reflector was to disappear, there
would be another tool in *weeks* on the market. Not as good as
Reflector, because it was of outstanding quality, but it would be
functional.

--- In altdotnet@..., "Sidar Ok" <sidarok@...> wrote:

>
> +1 on that.
>
> it is one of the 5 programs that has shortcuts on my desktop.
>
> It is also vital for me to learn usage of a new library / framework.
>
> On Fri, Aug 22, 2008 at 10:18 PM, Ayende Rahien <Ayende@...> wrote:
>
> >   you wouldn't believe how important Reflector is for my usual

workflow
> >
> > On Fri, Aug 22, 2008 at 5:14 PM, Gael Fraiteur <gael@...> wrote:
> >
> >> Yes, it's the field on which PostSharp and Cecil are nearly
> >> equivalent (competing?), Cecil with the advantage of *some*
analysis.
> >>
> >> Anyway - as long as Redgate makes it free, there is no market for
> >> such a tool -- and even OSS needs a market.
> >>
> >> It's amazing how Roeder managed to build a market on a tool that
is
> >> not only closed source, but with a license that automatically
expires
> >> after a number of days. It's unbelievable that he could have
people
> >> write plug-ins to that system...
> >>
> >> (Unbelievable? The reason is just that the tool is... a goodie,
not a
> >> tool that you vitally need to build your software. If Redgate
would
> >> simply supress Reflector tomorrow, nothing would happen. Excepted
> >> inconfort, maybe),
> >>
> >> --- In altdotnet@..., "Charlie Poole" <charlie@>
wrote:
> >> >
> >> > Yes, if I were starting such a project, I'd build on Cecil.
> >> >
> >> > Charlie
> >> >
> >> >
> >> >   _____
> >> >
> >> > From: altdotnet@...
[mailto:altdotnet@...]
> >> On Behalf
> >> > Of Ayende Rahien
> >> > Sent: Friday, August 22, 2008 6:58 AM
> >> > To: altdotnet@...
> >> > Subject: SPAM-LOW: Re: [altdotnet] Re: .NET Reflector now
belongs

> >> to Redgate
> >> >
> >> >
> >> > There is Cecil Flow Analsys, which can help
> >> >
> >> >
> >> > On Fri, Aug 22, 2008 at 9:56 AM, Gael Fraiteur <gael@> wrote:
> >> >
> >> >
> >> > You can make it one out of PostSharp, if you have the

courage :).
> >> >
> >> > You have a disassembler, an object model... you have to raise
the
> >> > level of abstraction from MSIL to C#, i.e. you have to write
the
> >> > decompiler.
> >> >
> >> > And that's not simple. I think Lutz had to hardcode all
patterns

> >> used
> >> > by well-known compilers...
> >> >
> >> > -gael
> >> >
> >> > --- In altdotnet@..., "Aaron Erickson"
> >> >
> >> > <aaron.c.erickson@> wrote:
> >> > >
> >> > > I sense an opportunity for an opportunist to step in and

provide a
> >> > > replacement OSS solution.
> >> > >
> >> > > On Fri, Aug 22, 2008 at 7:28 AM, Gabriel Schenker
> >> >
> >> > <gnschenker@>wrote:
> >> >
> >> > >
> >> > > >   after years of development Lutz Roeder obviously
abandons it's
> >> > flag
> >> > > > ship an has given (sold?) it to Redgate.
> >> > > > Although Redgate has annouced to provide a free version
also in
> >> > the
> >> > > > future I fear that in the long run you will have to pay
for the
> >> > full
> >> > > > version of Reflector.
> >> > > > Do we have to say goodbye to another indispensable OSS
tool?

> >> (see
> >> > e.g.
> >> > > > what happened to NCover and Testdriven.Net)
> >> > > >
> >> > > >
> >> > > >
> >> > >
> >> >
> >> >
> >> >
> >> >
> >> > ------------------------------------
> >> >
> >> > Yahoo! Groups Links
> >> >
> >>
> >>
> >>
> >> ------------------------------------
> >>
> >> Yahoo! Groups Links
> >>
> >>
> >>
> >>
> >  
> >
>
>
>
> --
> Sidar Ok
> http://www.sidarok.com
>

[Paulo Mouat-2]

On Sat, Aug 23, 2008 at 5:22 AM, Gael Fraiteur <gael@...> wrote:
> If Reflector was to disappear, there
> would be another tool in *weeks* on the market. Not as good as
> Reflector, because it was of outstanding quality, but it would be
> functional.

Try Decompiler.NET. Closed and commercial, but in some respects better
than Reflector--produces real, compilable, C# or VB code, including
the supporting solution and projects. I have no affiliation with
Decompiler.NET.

//p

[J Wynia-2]

On Sat, Aug 23, 2008 at 6:20 AM, Paulo Mouat <paulo.mouat@...> wrote:

>    Try Decompiler.NET. Closed and commercial, but in some respects better
> than Reflector--produces real, compilable, C# or VB code, including
> the supporting solution and projects. I have no affiliation with
> Decompiler.NET.
>






There are plugins that I've had installed in Reflector ever since I started
using it that generate compilable projects as well. I've used it on client
projects where they either had lost the source code that a previous
developer had built or their source code repository was corrupt, leaving me
with no source code to work with.

Being able to grab the DLL or EXE from the production server, decompile,
make changes and redeploy is critical to my consultancy. Paying a reasonable
cost is fine with me if that's what it takes to keep tools like this
available.

--
J Wynia
Software Consultant, Writer and Geek
Minneapolis, MN
j@...
"The glass isn't half full or half empty. It's just too big"
http://wynia.org

[Ayende Rahien-2]

Am I the only one who thinks that the work flow outlined here is WTF?

On Sat, Aug 23, 2008 at 9:48 AM, J Wynia <j@...> wrote:

>  On Sat, Aug 23, 2008 at 6:20 AM, Paulo Mouat <paulo.mouat@...>wrote:
>
>>    Try Decompiler.NET. Closed and commercial, but in some respects better
>> than Reflector--produces real, compilable, C# or VB code, including
>> the supporting solution and projects. I have no affiliation with
>> Decompiler.NET.
>>
>
>
>
>
>
>
> There are plugins that I've had installed in Reflector ever since I started
> using it that generate compilable projects as well. I've used it on client
> projects where they either had lost the source code that a previous
> developer had built or their source code repository was corrupt, leaving me
> with no source code to work with.
>
> Being able to grab the DLL or EXE from the production server, decompile,
> make changes and redeploy is critical to my consultancy. Paying a reasonable
> cost is fine with me if that's what it takes to keep tools like this
> available.
>
> --
> J Wynia
> Software Consultant, Writer and Geek
> Minneapolis, MN
> j@...
> "The glass isn't half full or half empty. It's just too big"
> http://wynia.org
>  
>

[Cory Foy-2]

Ayende Rahien wrote:
> Am I the only one who thinks that the work flow outlined here is WTF?

Probably not, but it makes sense to me. Sounds like he makes his living
with the customers who go, "Oh crap, our "star programmer" just left and
we have to get these fixes in, but he didn't leave any code." Probably
doing well for himself too.

--
Cory Foy
http://www.cornetdesign.com
http://www.agileflorida.com

[J Wynia-2]

On Sat, Aug 23, 2008 at 4:09 PM, Cory Foy <usergroup@...>wrote:

>   Ayende Rahien wrote:
> > Am I the only one who thinks that the work flow outlined here is WTF?
>
> Probably not, but it makes sense to me. Sounds like he makes his living
> with the customers who go, "Oh crap, our "star programmer" just left and
> we have to get these fixes in, but he didn't leave any code." Probably
> doing well for himself too.
>









Of course that workflow is WTF. There's a phrase from Yorkshire that I like,
"Where there's muck there's brass." Basically, there's a decent living to be
made where there's dirty work. And, WTF environments and WTF situations are
dirty work.

I know a lot of folks seek out gigs with startups and other greenfield
development, whre they get to start from scratch and build something new. It
makes sense. That's fun stuff. However, there's a lot more work out there in
messed up environments, working on messed up codebases, under less than
ideal conditions. And, there's much less competition for it.

The brownfield marketplace is thriving and growing every year as more and
more companies have custom software based on a mix of technology behind the
firewall, but none of the original authors anywhere in sight. They go to
most consultants and get quotes for rewriting their apps from scratch and,
when the quote comes in at $2+ million, they often drop the idea. Those of
us working in this space say, "Sure, we can make what you've got work for a
fraction of new".

I get to help clean up messes, face interesting challenges, make a decent
living working 4 days a week and have time to work on my own projects, open
source projects, etc. I enjoy what I do and market demand keeps me booked
solid. I look at it much like how the real money in the San Francisco gold
rush in the 19th century wasn't in gold. It was in blue jeans, pick axes and
shovels. Not nearly as sexy, but more reliable and steady. And, that fits my
pragmatic approach to life a lot more than the alternative.

--
J Wynia
Software Consultant, Writer and Geek
Minneapolis, MN
j@...
"The glass isn't half full or half empty. It's just too big"
http://wynia.org

[Ayende Rahien-2]

That is not the problem, I have been called into those situations. The
problem is that the description seems to exclude the part where you setup
explicit SCM practices.

On Sat, Aug 23, 2008 at 5:09 PM, Cory Foy <usergroup@...>wrote:

> Ayende Rahien wrote:
> > Am I the only one who thinks that the work flow outlined here is WTF?
>
> Probably not, but it makes sense to me. Sounds like he makes his living
> with the customers who go, "Oh crap, our "star programmer" just left and
> we have to get these fixes in, but he didn't leave any code." Probably
> doing well for himself too.
>
> --
> Cory Foy
> http://www.cornetdesign.com
> http://www.agileflorida.com
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>

[Cory Foy-2]

You mean having a file share isn't an explicit SCM practice? ;)

So yes, to having to have people who go in and do this, it is a total WTF.

Cory

Ayende Rahien wrote:

> That is not the problem, I have been called into those situations. The
> problem is that the description seems to exclude the part where you setup
> explicit SCM practices.
>
> On Sat, Aug 23, 2008 at 5:09 PM, Cory Foy <usergroup@...>wrote:
>
>> Ayende Rahien wrote:
>>> Am I the only one who thinks that the work flow outlined here is WTF?
>> Probably not, but it makes sense to me. Sounds like he makes his living
>> with the customers who go, "Oh crap, our "star programmer" just left and
>> we have to get these fixes in, but he didn't leave any code." Probably
>> doing well for himself too.
>>
>> --
>> Cory Foy
>> http://www.cornetdesign.com
>> http://www.agileflorida.com
>>
>>
>> ------------------------------------
>>
>> Yahoo! Groups Links
>>
>>
>>
>>
>

--
Cory Foy
http://www.cornetdesign.com
http://www.agileflorida.com

[J Wynia-2]

Lets step back and look at the sequence of events here.

1. Someone pointed out an alternate piece of software to Reflector,
indicating that it supported exporting project and solution files from the
assembly.
2. I, knowing that there was a plugin for Reflector that does the same
thing, pointed that out.
3. Given that I use that particular feature more often than I'd really like,
but always in situations where I have no other choice, I wrote a 22 word
aside that expressed my reliance on the feature in those situations,
followed by an indication that I'd pay for it.
4. (This is an implied event) Several people on this list came to the
conclusion that my 22 word statement described the entire breadth and depth
of my workflow, even in these situations.
5. One of them asked whether my workflow was "WTF".
6. One of the list members who appears not to have been in the group from #4
clarified what he thought I meant.
7. Agreeing with the person from #6 on what I meant, I attempted to describe
the situations where this is the case.
8. The person from #5 said they didn't mean the situations described were
WTF, but the 22 word workflow description was, based upon intense scrutiny,
which revealed that the statement was lacking a description of my source
control management in these situations.
9. The statement below was posted, clearly indicating that the implied group
in #4 exists and consists of more than one member, furthering the
implication that I'm some sort of knuckle-dragging moron who would consider
a shared drive a source control management system.
10. I find myself lving out this web comic in an attempt to set the record
straight. - http://xkcd.com/386/

Had I known that my offhand 22 word description of my Reflector-based
workflow would be used as the complete picture of how I work, I would
probably just omitted it altogether. However, it's out there, and thanks to
the beauty of the Internet, attached to my name in a community of people
that I respect, so I'd like to set the record straight.

In these situations where I need to use Reflector, here's exactly what
happens.

I take a copy of the assembly from production, one from staging, and one
from dev. I also take a copy of the source from every possible location.
This latest time, that meant a copy from SourceSafe, a copy from Subversion
and a copy from a former developer's workstation.

Those copies in source form are built into the appropriate assemblies.

All 6 assemblies are run through Reflector to generate project and solution
files, into separate directories, named according to the source. The
decompiled assemblies are used even when I have the source because you're
comparing apples to apples for actual differences in the output assemblies.

Using WinMerge, I work through the permutations between all 6 directories,
generating a list of all of the differences. I use those diffs to figure out
and reconstruct a history. That history is put into a clean Subversion
repository in the appropriate order.

Then I look at how the production snapshot in particular fits into the
equation. That nearly always results in a series of meetings with the client
to determine how those difference fit with what they intended.

Only once that's all straightened out, does any actual work begin on that
codebase. Typically, at that point, I start using Git for local versioning
and pushing substantive changes up to Subversion or whatever the client
prefers. My Git commits are a combination of deliberate commits when I want
a snapshot and scripted commits as part of my post-build events in Visual
Studio. In other words, a typical day results in nearly constant commits and
the ability to track ALL of the changes.

When code is ready for deployment, it's branched into a release snapshot in
the repository and the deployment is done straight out of the repository.

So, when I neglected to include 3 words about my source code management, of
course, rather than assume that I actually do the professional thing and go
above and beyond to keep things in order, it was easy to assume that I must
be a bumbling novice and make a joke. Wrong, but easy.

Please, do future conversations a favor and give people the benefit of the
doubt. I'm doing that to those of you who fell into the group in #4 and am
going to assume that you just made a quick glance at my post and, since it
sounded like something you'd heard from someone who DID skip the SCM portion
of the lifecycle, figured I was in the same category. I'll assume that
instead of assuming that rather than much less charitable interpretations.


On Sat, Aug 23, 2008 at 10:15 PM, Cory Foy <usergroup@...>wrote:

>   You mean having a file share isn't an explicit SCM practice? ;)
>
> So yes, to having to have people who go in and do this, it is a total WTF.
>
> Cory
>
>
> Ayende Rahien wrote:
> > That is not the problem, I have been called into those situations. The
> > problem is that the description seems to exclude the part where you setup
> > explicit SCM practices.
>

--
J Wynia
Software Consultant, Writer and Geek
Minneapolis, MN
j@...
"The glass isn't half full or half empty. It's just too big"
http://wynia.org

[Alan Dean]

Stepping back a little ...

After you were called out and have replied, we have a much clearer
picture of what you actually do.

I suspect that this group has many more lurkers than participants
(this is not uncommon). What is more, I suspect that many of the
lurkers are those who are less experienced (at least in terms of
alt.net) and who thus don't feel competent to engage.

Such people might have gone away with an impression that the
'shorthand' version of your workflow originally posted would be a good
thing to do.

Now, they know better :-)

What is more, I wouldn't consider myself inexperienced - but I learnt
something. Namely, when (in extremis) you might need to introduce a
delta anywhere near Live - you can use Git to mitigate the
(horrendous) risks.

Thankfully, I haven't had a problem both so severe and so
time-sensitive in many years but I gained some knowledge - so I thank
you!

Alan

On Sun, Aug 24, 2008 at 7:20 PM, J Wynia <j@...> wrote:

> Lets step back and look at the sequence of events here.
>
> 1. Someone pointed out an alternate piece of software to Reflector,
> indicating that it supported exporting project and solution files from the
> assembly.
> 2. I, knowing that there was a plugin for Reflector that does the same
> thing, pointed that out.
> 3. Given that I use that particular feature more often than I'd really like,
> but always in situations where I have no other choice, I wrote a 22 word
> aside that expressed my reliance on the feature in those situations,
> followed by an indication that I'd pay for it.
> 4. (This is an implied event) Several people on this list came to the
> conclusion that my 22 word statement described the entire breadth and depth
> of my workflow, even in these situations.
> 5. One of them asked whether my workflow was "WTF".
> 6. One of the list members who appears not to have been in the group from #4
> clarified what he thought I meant.
> 7. Agreeing with the person from #6 on what I meant, I attempted to describe
> the situations where this is the case.
> 8. The person from #5 said they didn't mean the situations described were
> WTF, but the 22 word workflow description was, based upon intense scrutiny,
> which revealed that the statement was lacking a description of my source
> control management in these situations.
> 9. The statement below was posted, clearly indicating that the implied group
> in #4 exists and consists of more than one member, furthering the
> implication that I'm some sort of knuckle-dragging moron who would consider
> a shared drive a source control management system.
> 10. I find myself lving out this web comic in an attempt to set the record
> straight. - http://xkcd.com/386/
>
> Had I known that my offhand 22 word description of my Reflector-based
> workflow would be used as the complete picture of how I work, I would
> probably just omitted it altogether. However, it's out there, and thanks to
> the beauty of the Internet, attached to my name in a community of people
> that I respect, so I'd like to set the record straight.
>
> In these situations where I need to use Reflector, here's exactly what
> happens.
>
> I take a copy of the assembly from production, one from staging, and one
> from dev. I also take a copy of the source from every possible location.
> This latest time, that meant a copy from SourceSafe, a copy from Subversion
> and a copy from a former developer's workstation.
>
> Those copies in source form are built into the appropriate assemblies.
>
> All 6 assemblies are run through Reflector to generate project and solution
> files, into separate directories, named according to the source. The
> decompiled assemblies are used even when I have the source because you're
> comparing apples to apples for actual differences in the output assemblies.
>
> Using WinMerge, I work through the permutations between all 6 directories,
> generating a list of all of the differences. I use those diffs to figure out
> and reconstruct a history. That history is put into a clean Subversion
> repository in the appropriate order.
>
> Then I look at how the production snapshot in particular fits into the
> equation. That nearly always results in a series of meetings with the client
> to determine how those difference fit with what they intended.
>
> Only once that's all straightened out, does any actual work begin on that
> codebase. Typically, at that point, I start using Git for local versioning
> and pushing substantive changes up to Subversion or whatever the client
> prefers. My Git commits are a combination of deliberate commits when I want
> a snapshot and scripted commits as part of my post-build events in Visual
> Studio. In other words, a typical day results in nearly constant commits and
> the ability to track ALL of the changes.
>
> When code is ready for deployment, it's branched into a release snapshot in
> the repository and the deployment is done straight out of the repository.
>
> So, when I neglected to include 3 words about my source code management, of
> course, rather than assume that I actually do the professional thing and go
> above and beyond to keep things in order, it was easy to assume that I must
> be a bumbling novice and make a joke. Wrong, but easy.
>
> Please, do future conversations a favor and give people the benefit of the
> doubt. I'm doing that to those of you who fell into the group in #4 and am
> going to assume that you just made a quick glance at my post and, since it
> sounded like something you'd heard from someone who DID skip the SCM portion
> of the lifecycle, figured I was in the same category. I'll assume that
> instead of assuming that rather than much less charitable interpretations.
>
>
> On Sat, Aug 23, 2008 at 10:15 PM, Cory Foy <usergroup@...>
> wrote:
>>
>> You mean having a file share isn't an explicit SCM practice? ;)
>>
>> So yes, to having to have people who go in and do this, it is a total WTF.
>>
>> Cory
>>
>> Ayende Rahien wrote:
>> > That is not the problem, I have been called into those situations. The
>> > problem is that the description seems to exclude the part where you
>> > setup
>> > explicit SCM practices.

[J Wynia-2]

Everyone has something to teach and everyone has something they can learn.
Unfortunately, inflammatory phrases like "WTF" and sarcastic insults rarely
create an environment that encourages that kind of exchange.

I am 100% in favor of the sharing of information. That's why I put as much
stuff up on my website as I do.

Had someone asked me to clarify that workflow, I'd have done so happily.
Often, asking exactly such a question allows both people to clarify the bits
that were confusing to them. This is the kind of stuff you don't often find
in books or classes. I'm glad you found it useful.


On Sun, Aug 24, 2008 at 3:30 PM, Alan Dean <alan.dean@...> wrote:

>   Stepping back a little ...
>
> After you were called out and have replied, we have a much clearer
> picture of what you actually do.
>
> I suspect that this group has many more lurkers than participants
> (this is not uncommon). What is more, I suspect that many of the
> lurkers are those who are less experienced (at least in terms of
> alt.net) and who thus don't feel competent to engage.
>
> Such people might have gone away with an impression that the
> 'shorthand' version of your workflow originally posted would be a good
> thing to do.
>
> Now, they know better :-)
>
> What is more, I wouldn't consider myself inexperienced - but I learnt
> something. Namely, when (in extremis) you might need to introduce a
> delta anywhere near Live - you can use Git to mitigate the
> (horrendous) risks.
>
> Thankfully, I haven't had a problem both so severe and so
> time-sensitive in many years but I gained some knowledge - so I thank
> you!
>
> Alan
>
>
> On Sun, Aug 24, 2008 at 7:20 PM, J Wynia <j@... <j%40wynia.org>>
> wrote:
> > Lets step back and look at the sequence of events here.
> >
> > 1. Someone pointed out an alternate piece of software to Reflector,
> > indicating that it supported exporting project and solution files from
> the
> > assembly.
> > 2. I, knowing that there was a plugin for Reflector that does the same
> > thing, pointed that out.
> > 3. Given that I use that particular feature more often than I'd really
> like,
> > but always in situations where I have no other choice, I wrote a 22 word
> > aside that expressed my reliance on the feature in those situations,
> > followed by an indication that I'd pay for it.
> > 4. (This is an implied event) Several people on this list came to the
> > conclusion that my 22 word statement described the entire breadth and
> depth
> > of my workflow, even in these situations.
> > 5. One of them asked whether my workflow was "WTF".
> > 6. One of the list members who appears not to have been in the group from
> #4
> > clarified what he thought I meant.
> > 7. Agreeing with the person from #6 on what I meant, I attempted to
> describe
> > the situations where this is the case.
> > 8. The person from #5 said they didn't mean the situations described were
> > WTF, but the 22 word workflow description was, based upon intense
> scrutiny,
> > which revealed that the statement was lacking a description of my source
> > control management in these situations.
> > 9. The statement below was posted, clearly indicating that the implied
> group
> > in #4 exists and consists of more than one member, furthering the
> > implication that I'm some sort of knuckle-dragging moron who would
> consider
> > a shared drive a source control management system.
> > 10. I find myself lving out this web comic in an attempt to set the
> record
> > straight. - http://xkcd.com/386/
> >
> > Had I known that my offhand 22 word description of my Reflector-based
> > workflow would be used as the complete picture of how I work, I would
> > probably just omitted it altogether. However, it's out there, and thanks
> to
> > the beauty of the Internet, attached to my name in a community of people
> > that I respect, so I'd like to set the record straight.
> >
> > In these situations where I need to use Reflector, here's exactly what
> > happens.
> >
> > I take a copy of the assembly from production, one from staging, and one
> > from dev. I also take a copy of the source from every possible location.
> > This latest time, that meant a copy from SourceSafe, a copy from
> Subversion
> > and a copy from a former developer's workstation.
> >
> > Those copies in source form are built into the appropriate assemblies.
> >
> > All 6 assemblies are run through Reflector to generate project and
> solution
> > files, into separate directories, named according to the source. The
> > decompiled assemblies are used even when I have the source because you're
> > comparing apples to apples for actual differences in the output
> assemblies.
> >
> > Using WinMerge, I work through the permutations between all 6
> directories,
> > generating a list of all of the differences. I use those diffs to figure
> out
> > and reconstruct a history. That history is put into a clean Subversion
> > repository in the appropriate order.
> >
> > Then I look at how the production snapshot in particular fits into the
> > equation. That nearly always results in a series of meetings with the
> client
> > to determine how those difference fit with what they intended.
> >
> > Only once that's all straightened out, does any actual work begin on that
> > codebase. Typically, at that point, I start using Git for local
> versioning
> > and pushing substantive changes up to Subversion or whatever the client
> > prefers. My Git commits are a combination of deliberate commits when I
> want
> > a snapshot and scripted commits as part of my post-build events in Visual
> > Studio. In other words, a typical day results in nearly constant commits
> and
> > the ability to track ALL of the changes.
> >
> > When code is ready for deployment, it's branched into a release snapshot
> in
> > the repository and the deployment is done straight out of the repository.
> >
> > So, when I neglected to include 3 words about my source code management,
> of
> > course, rather than assume that I actually do the professional thing and
> go
> > above and beyond to keep things in order, it was easy to assume that I
> must
> > be a bumbling novice and make a joke. Wrong, but easy.
> >
> > Please, do future conversations a favor and give people the benefit of
> the
> > doubt. I'm doing that to those of you who fell into the group in #4 and
> am
> > going to assume that you just made a quick glance at my post and, since
> it
> > sounded like something you'd heard from someone who DID skip the SCM
> portion
> > of the lifecycle, figured I was in the same category. I'll assume that
> > instead of assuming that rather than much less charitable
> interpretations.
> >
> >
> > On Sat, Aug 23, 2008 at 10:15 PM, Cory Foy <usergroup@...<usergroup%40cornetdesign.com>
> >
> > wrote:
> >>
> >> You mean having a file share isn't an explicit SCM practice? ;)
> >>
> >> So yes, to having to have people who go in and do this, it is a total
> WTF.
> >>
> >> Cory
> >>
> >> Ayende Rahien wrote:
> >> > That is not the problem, I have been called into those situations. The
> >> > problem is that the description seems to exclude the part where you
> >> > setup
> >> > explicit SCM practices.
>  
>

--
J Wynia
Software Consultant, Writer and Geek
Minneapolis, MN
j@...
"The glass isn't half full or half empty. It's just too big"
http://wynia.org

[Ayende Rahien-2]

J,First, allow me to apologize.
Second, the reason that I reacted this way was seeing a client that
did not move
beyond the decompilation phase.

I have a project, to this day, that not only compiles the source code, but
also embeds the entire project (files, solution, build scripts) into the
output assembly. That was I _know_ they can't lose the code.


On Sun, Aug 24, 2008 at 5:31 PM, J Wynia <j@...> wrote:

>  Everyone has something to teach and everyone has something they can
> learn. Unfortunately, inflammatory phrases like "WTF" and sarcastic insults
> rarely create an environment that encourages that kind of exchange.
>
> I am 100% in favor of the sharing of information. That's why I put as much
> stuff up on my website as I do.
>
> Had someone asked me to clarify that workflow, I'd have done so happily.
> Often, asking exactly such a question allows both people to clarify the bits
> that were confusing to them. This is the kind of stuff you don't often find
> in books or classes. I'm glad you found it useful.
>
>
>
> On Sun, Aug 24, 2008 at 3:30 PM, Alan Dean <alan.dean@...> wrote:
>
>>   Stepping back a little ...
>>
>> After you were called out and have replied, we have a much clearer
>> picture of what you actually do.
>>
>> I suspect that this group has many more lurkers than participants
>> (this is not uncommon). What is more, I suspect that many of the
>> lurkers are those who are less experienced (at least in terms of
>> alt.net) and who thus don't feel competent to engage.
>>
>> Such people might have gone away with an impression that the
>> 'shorthand' version of your workflow originally posted would be a good
>> thing to do.
>>
>> Now, they know better :-)
>>
>> What is more, I wouldn't consider myself inexperienced - but I learnt
>> something. Namely, when (in extremis) you might need to introduce a
>> delta anywhere near Live - you can use Git to mitigate the
>> (horrendous) risks.
>>
>> Thankfully, I haven't had a problem both so severe and so
>> time-sensitive in many years but I gained some knowledge - so I thank
>> you!
>>
>> Alan
>>
>>
>> On Sun, Aug 24, 2008 at 7:20 PM, J Wynia <j@... <j%40wynia.org>>
>> wrote:
>> > Lets step back and look at the sequence of events here.
>> >
>> > 1. Someone pointed out an alternate piece of software to Reflector,
>> > indicating that it supported exporting project and solution files from
>> the
>> > assembly.
>> > 2. I, knowing that there was a plugin for Reflector that does the same
>> > thing, pointed that out.
>> > 3. Given that I use that particular feature more often than I'd really
>> like,
>> > but always in situations where I have no other choice, I wrote a 22 word
>> > aside that expressed my reliance on the feature in those situations,
>> > followed by an indication that I'd pay for it.
>> > 4. (This is an implied event) Several people on this list came to the
>> > conclusion that my 22 word statement described the entire breadth and
>> depth
>> > of my workflow, even in these situations.
>> > 5. One of them asked whether my workflow was "WTF".
>> > 6. One of the list members who appears not to have been in the group
>> from #4
>> > clarified what he thought I meant.
>> > 7. Agreeing with the person from #6 on what I meant, I attempted to
>> describe
>> > the situations where this is the case.
>> > 8. The person from #5 said they didn't mean the situations described
>> were
>> > WTF, but the 22 word workflow description was, based upon intense
>> scrutiny,
>> > which revealed that the statement was lacking a description of my source
>> > control management in these situations.
>> > 9. The statement below was posted, clearly indicating that the implied
>> group
>> > in #4 exists and consists of more than one member, furthering the
>> > implication that I'm some sort of knuckle-dragging moron who would
>> consider
>> > a shared drive a source control management system.
>> > 10. I find myself lving out this web comic in an attempt to set the
>> record
>> > straight. - http://xkcd.com/386/
>> >
>> > Had I known that my offhand 22 word description of my Reflector-based
>> > workflow would be used as the complete picture of how I work, I would
>> > probably just omitted it altogether. However, it's out there, and thanks
>> to
>> > the beauty of the Internet, attached to my name in a community of people
>> > that I respect, so I'd like to set the record straight.
>> >
>> > In these situations where I need to use Reflector, here's exactly what
>> > happens.
>> >
>> > I take a copy of the assembly from production, one from staging, and one
>> > from dev. I also take a copy of the source from every possible location.
>> > This latest time, that meant a copy from SourceSafe, a copy from
>> Subversion
>> > and a copy from a former developer's workstation.
>> >
>> > Those copies in source form are built into the appropriate assemblies.
>> >
>> > All 6 assemblies are run through Reflector to generate project and
>> solution
>> > files, into separate directories, named according to the source. The
>> > decompiled assemblies are used even when I have the source because
>> you're
>> > comparing apples to apples for actual differences in the output
>> assemblies.
>> >
>> > Using WinMerge, I work through the permutations between all 6
>> directories,
>> > generating a list of all of the differences. I use those diffs to figure
>> out
>> > and reconstruct a history. That history is put into a clean Subversion
>> > repository in the appropriate order.
>> >
>> > Then I look at how the production snapshot in particular fits into the
>> > equation. That nearly always results in a series of meetings with the
>> client
>> > to determine how those difference fit with what they intended.
>> >
>> > Only once that's all straightened out, does any actual work begin on
>> that
>> > codebase. Typically, at that point, I start using Git for local
>> versioning
>> > and pushing substantive changes up to Subversion or whatever the client
>> > prefers. My Git commits are a combination of deliberate commits when I
>> want
>> > a snapshot and scripted commits as part of my post-build events in
>> Visual
>> > Studio. In other words, a typical day results in nearly constant commits
>> and
>> > the ability to track ALL of the changes.
>> >
>> > When code is ready for deployment, it's branched into a release snapshot
>> in
>> > the repository and the deployment is done straight out of the
>> repository.
>> >
>> > So, when I neglected to include 3 words about my source code management,
>> of
>> > course, rather than assume that I actually do the professional thing and
>> go
>> > above and beyond to keep things in order, it was easy to assume that I
>> must
>> > be a bumbling novice and make a joke. Wrong, but easy.
>> >
>> > Please, do future conversations a favor and give people the benefit of
>> the
>> > doubt. I'm doing that to those of you who fell into the group in #4 and
>> am
>> > going to assume that you just made a quick glance at my post and, since
>> it
>> > sounded like something you'd heard from someone who DID skip the SCM
>> portion
>> > of the lifecycle, figured I was in the same category. I'll assume that
>> > instead of assuming that rather than much less charitable
>> interpretations.
>> >
>> >
>> > On Sat, Aug 23, 2008 at 10:15 PM, Cory Foy <usergroup@...<usergroup%40cornetdesign.com>
>> >
>> > wrote:
>> >>
>> >> You mean having a file share isn't an explicit SCM practice? ;)
>> >>
>> >> So yes, to having to have people who go in and do this, it is a total
>> WTF.
>> >>
>> >> Cory
>> >>
>> >> Ayende Rahien wrote:
>> >> > That is not the problem, I have been called into those situations.
>> The
>> >> > problem is that the description seems to exclude the part where you
>> >> > setup
>> >> > explicit SCM practices.
>>
>
>
>
> --
> J Wynia
> Software Consultant, Writer and Geek
> Minneapolis, MN
> j@...
> "The glass isn't half full or half empty. It's just too big"
> http://wynia.org
>  
>

[Cory Foy-2]

Hi J,

J Wynia wrote:

> 9. The statement below was posted, clearly indicating that the implied group
> in #4 exists and consists of more than one member, furthering the
> implication that I'm some sort of knuckle-dragging moron who would consider
> a shared drive a source control management system.
> 10. I find myself lving out this web comic in an attempt to set the record
> straight. - http://xkcd.com/386/
>
> Had I known that my offhand 22 word description of my Reflector-based
> workflow would be used as the complete picture of how I work, I would
> probably just omitted it altogether. However, it's out there, and thanks to
> the beauty of the Internet, attached to my name in a community of people
> that I respect, so I'd like to set the record straight.

Thank you for setting the record straight. However, I apologize for you
thinking that Ayende and I were actually referring to *you* and *your
situation* about WTFs. Well, Ayende might have, I'm not going to put
words in his mouth.

In general, the /necessity/ of the workflow you describe is the WTF that
at least I was thinking about. I envisioned a client which had someone
leave, and could bring someone like you aboard to help get them
straightened out and back on the good path. If anything, I though highly
of the work you did, because it ain't easy.

I did somehow miss that Ayende's comment about setting up explicit SCM
practices could be construed as pointing to you specifically. I, again,
was thinking of the fictional client that you, the superhero who could
sweep in and fix them up, would be extinct because of if it was in place.

I don't think that you needed to provide any other information that what
you did. I should have thought clearer about the impact the conversation
was having. Instead, I was sparring with Ayende about a fictional
company I had in my head who had a need to have someone like you come in
and save them. Brownfield indeed.

--
Cory Foy
http://www.cornetdesign.com
http://www.agileflorida.com

[josh robb-2]

On Mon, Aug 25, 2008 at 12:33 PM, Ayende Rahien <Ayende@...> wrote:
> I have a project, to this day, that not only compiles the source code, but
> also embeds the entire project (files, solution, build scripts) into the
> output assembly. That was I _know_ they can't lose the code.

Ayende - That's awesome - I _wish_ I'd thought of that when I was
working for a previous client!

Embeded Resource FTW!

[Ayende Rahien-2]

It is something like this:
http://www.ayende.com/Blog/archive/2006/10/09/7207.aspx

On Mon, Aug 25, 2008 at 5:02 AM, josh robb <josh.robb@...> wrote:

> On Mon, Aug 25, 2008 at 12:33 PM, Ayende Rahien <Ayende@...> wrote:
> > I have a project, to this day, that not only compiles the source code,
> but
> > also embeds the entire project (files, solution, build scripts) into the
> > output assembly. That was I _know_ they can't lose the code.
>
> Ayende - That's awesome - I _wish_ I'd thought of that when I was
> working for a previous client!
>
> Embeded Resource FTW!
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>

[Alan Dean]

Is that what they meant when they were talking about asexual
reproduction in Biology class??

On Mon, Aug 25, 2008 at 11:40 AM, Ayende Rahien <Ayende@...> wrote:

> It is something like this:
> http://www.ayende.com/Blog/archive/2006/10/09/7207.aspx
>
> On Mon, Aug 25, 2008 at 5:02 AM, josh robb <josh.robb@...> wrote:
>>
>> On Mon, Aug 25, 2008 at 12:33 PM, Ayende Rahien <Ayende@...> wrote:
>> > I have a project, to this day, that not only compiles the source code,
>> > but
>> > also embeds the entire project (files, solution, build scripts) into the
>> > output assembly. That was I _know_ they can't lose the code.
>>
>> Ayende - That's awesome - I _wish_ I'd thought of that when I was
>> working for a previous client!
>>
>> Embeded Resource FTW!

[J Wynia-2]

On Sun, Aug 24, 2008 at 10:16 PM, Cory Foy <usergroup@...>wrote:

>   Hi J,
> Thank you for setting the record straight. However, I apologize for you
> thinking that Ayende and I were actually referring to *you* and *your
> situation* about WTFs. Well, Ayende might have, I'm not going to put
> words in his mouth.
>
> In general, the /necessity/ of the workflow you describe is the WTF that
> at least I was thinking about. I envisioned a client which had someone
> leave, and could bring someone like you aboard to help get them
> straightened out and back on the good path. If anything, I though highly
> of the work you did, because it ain't easy.
>
> I did somehow miss that Ayende's comment about setting up explicit SCM
> practices could be construed as pointing to you specifically. I, again,
> was thinking of the fictional client that you, the superhero who could
> sweep in and fix them up, would be extinct because of if it was in place.
>
> I don't think that you needed to provide any other information that what
> you did. I should have thought clearer about the impact the conversation
> was having. Instead, I was sparring with Ayende about a fictional
> company I had in my head who had a need to have someone like you come in
> and save them. Brownfield indeed.
>
>

No problem. Take a little ambiguous communication, complicated issues and my
predisposition this weekend to be a little defensive (there was an envelope
from the IRS about some stuff my business partner didn't take care of
propertly tacked to my door on Friday) and I probably overreacted a bit as
well.

--
J Wynia
Software Consultant, Writer and Geek
Minneapolis, MN
j@...
"The glass isn't half full or half empty. It's just too big"
http://wynia.org

[sidarok]

Would any one be interested on an open source project on this ?
(contributing - using)

On Mon, Aug 25, 2008 at 3:59 PM, J Wynia <j@...> wrote:

>    On Sun, Aug 24, 2008 at 10:16 PM, Cory Foy <usergroup@...>wrote:
>
>>   Hi J,
>> Thank you for setting the record straight. However, I apologize for you
>> thinking that Ayende and I were actually referring to *you* and *your
>> situation* about WTFs. Well, Ayende might have, I'm not going to put
>> words in his mouth.
>>
>> In general, the /necessity/ of the workflow you describe is the WTF that
>> at least I was thinking about. I envisioned a client which had someone
>> leave, and could bring someone like you aboard to help get them
>> straightened out and back on the good path. If anything, I though highly
>> of the work you did, because it ain't easy.
>>
>> I did somehow miss that Ayende's comment about setting up explicit SCM
>> practices could be construed as pointing to you specifically. I, again,
>> was thinking of the fictional client that you, the superhero who could
>> sweep in and fix them up, would be extinct because of if it was in place.
>>
>> I don't think that you needed to provide any other information that what
>> you did. I should have thought clearer about the impact the conversation
>> was having. Instead, I was sparring with Ayende about a fictional
>> company I had in my head who had a need to have someone like you come in
>> and save them. Brownfield indeed.
>>
>>
>
> No problem. Take a little ambiguous communication, complicated issues and
> my predisposition this weekend to be a little defensive (there was an
> envelope from the IRS about some stuff my business partner didn't take care
> of propertly tacked to my door on Friday) and I probably overreacted a bit
> as well.
>
> --
> J Wynia
> Software Consultant, Writer and Geek
> Minneapolis, MN
> j@...
> "The glass isn't half full or half empty. It's just too big"
> http://wynia.org
>
>
>

--
Sidar Ok
http://www.sidarok.com