Re: [review][LEAF] Review of LEAF starts today : May 22 - May 31

From: pbristow_at_[hidden]
Date: 2020-06-02 16:48:31

> -----Original Message-----
> From: Boost-users <boost-users-bounces_at_[hidden]> On Behalf Of Michael Caisse via Boost-users
> Sent: 22 May 2020 10:32
> To: Boost users list <boost-users_at_[hidden]>
> Cc: Michael Caisse <mcaisse-lists_at_[hidden]>
> Subject: [Boost-users] [review][LEAF] Review of LEAF starts today : May 22 - May 31
> 
> The Boost formal review of Emil Dotchevski's LEAF (Lightweight Error Augmentation Framework) library
> will take place from May 22 through May 31st.
> 
> LEAF isn't just another error reporting library in the family of expected-like types. It provides a unique
> take on error handling which plays into usability, flexibility, and performance.
> 
> Some other questions you might want to consider answering:
> 
>   - What is your evaluation of the design?

Above my pay-grade, but looks reasonable.

>   - What is your evaluation of the implementation?

Above my pay-grade, but looks as though it works OK from examples and user experience.  Feels refined and the product of much work.

>   - What is your evaluation of the documentation?

Outstanding.  Excellent.

>   - What is your evaluation of the potential usefulness of the library?

I'm sure some people will find it useful.

>   - Did you try to use the library? 

I didn't try to use it - because I am not too unhappy with 'simple' try'n'catch exception programming.

>   - How much effort did you put into your evaluation? 

I've read, and re-read the documentation and examples, and all the reviews.

>   - Are you knowledgeable about the problem domain?

No - but the Big Thing I know is that the Less-expected Doesn't Work Well.
<rant>
* The hardware seems little help - bounds like containers are only available slowly and painfully.
* The software seems to have to be either fast and unhelpful, or slightly helpful at major computational cost (C++ exceptions).
* The programmers are idle and unhelpful.  How my blood pressure has been raised by shouting "Which file?" to "File not found."!
* Too much give-up quickly  "Something went wrong." Call C++ std::terminate ☹ 
* OK - enough - everyone agrees that things are not good.
</rant>

LEAF isn't The Silver Bullet, but clearly is reasonable way of doing things, and well done.

So the problem for Boost is whether to reject it on grounds that better ways exists, or accept it as a useful tool, helpfully documented.

It raises the number of ways for doing these things from perhaps three to four, but that's not much worse - it just is confusing.
(I am not over concerned at Boost Bloat - it is only programmers disk space that is idle/wasted. We MUST sell that better!)

So in the end I conclude that it turns on deciding if Boost should offer a choice tools, or should it make a single recommendation?

I see no case for any one Best Exceptions Tool, so I believe that we should ACCEPT LEAF.

Paul A. Bristow