Python 3 put-downs: What's the point?
in [Python]
|
Prev: Getting the name of the file that imported current module
Next: VICIOUS Enemy Insurgents kill 34 US soldiers after 9 hrs attack - Bellies ripped open and numerous injured - call for TRUE PATRIOTS . The deep undercover spies mislead the public on the facts
From: Tim Chase on 5 Jul 2010 15:32 On 07/05/2010 02:50 AM, Gregor Horvath wrote: > Am Sun, 04 Jul 2010 18:51:54 -0500 > schrieb Tim Chase<python.list(a)tim.thechases.com>: > >> I think it's the same venting of frustration that caused veteran >> VB6 developers to start calling VB.Net "Visual Fred" -- the >> language was too different and too non-backwards-compatible. >> > > VB6 -> VB.NET and Python 2 -> 3 is not a valid comparison. > > VB6 and VB.NET are totally different languages and technologies, with > some similarity in syntax. This is not true for Python 2->3. > This is an healthy organic language growth, not an abandon of a > language. The quintessential example is Py3's breaking of Hello World. It's a spectrum of language changes -- Visual Fred just happens to be MUCH further down the same spectrum having more dramatic changes. Only a subset of $OLD_VER (whether Py2 or VB6) code will run unmodified under $NEW_VER (whether Py3 or VB.Net). It just happens that the subset for Python is considerably larger than the subset for VB (and Python's conversion tools seem a little more useful than VB's, IMHO). IIRC, neither raw VB6 nor Py2 byte-code will run raw in the new environment (old VB .exe files don't make use of .Net libraries/CLR, nor do Py2 .pyc files run under Py3) so a project-rebuild is a minimum (though in Py3, s/minimum/negligible/) requirement. A little defensive coding in $OLD_VER also helps, and here I'd say Python developers had a MUCH longer lead-time to understand scope & magnitude of the coming changes; VB6 developers (former self included) had VB.Net foisted on them with much less heralding about the areas-of-breakage. I'm very much +0 on Py3...it doesn't impact my life yet and it's not a regular part of my coding, but the changes I've seen are good for the language and the future of Python. But breaking-changes freak some folks out, leading to the put-downs referenced by the OP. As a former VB6 developer, the shift to VB.Net was enough to send me packing. The shift from Py2 to Py3 will be bumpy, but not enough to lose me as a developer. -tkc
From: CM on 5 Jul 2010 16:41 On Jul 5, 2:33 pm, Terry Reedy <tjre...(a)udel.edu> wrote: > On 7/4/2010 9:20 PM, CM wrote: > > > On Jul 4, 7:14 pm, Terry Reedy<tjre...(a)udel.edu> wrote: > > > I think there's a good point to Python 3 put-downs (if I take put-down > > to mean generally reasonable criticism, which is what I've read here > > recently, and not trolling). And that is simply to register > > dissent. > > But dissent from what? > > Dissent from something obviously true? > (like 'Pythonx.y is useful to some people') > > Dissent from something obvious false, that no one has said? > (like 'Everyone should switch to Pythonx.y') I was thinking more like dissent from something that is not obviously true or false, but a matter of debate, like some of the decisions behind Python 3 itself or how the transition is being managed. I got the sense that was about where the complaints lie. Some of the responses to those complaints were educational to me, so I didn't mind reading the exchanges. > > Any online group is an opportunity to register dissent in a way that > > is public, open, immediate, interactive, and will (probably) be > > preserved for historians to check. The fact is, some people have > > gripes with Python 3; they are letting it be known. > > I have several 'gripes' with 2.7 and it is currently useless to me. > Should I let them be known? How many times? Maybe you should; maybe it can be constructive criticism to developers or can jog someone to tell you something that you didn't know. How many times? Once, maybe twice. I agree one can overdo it, and maybe you've read more of the gripes than I have and it seems repetitive by now. > > If no one did, > > there could be no later time at which people could look back and know > > what the reaction was to its introduction--it would just be a blank. > > Aren't opinions that dissent from the prevailing ones important to > > register, whether one thinks they are right or wrong? > > Do you agree with me that the same criteria for gripe legitimacy should > be applied equally to all Python versions (even if we should disagree on > what those criteria should be)? I think so, sure.
From: Steven D'Aprano on 5 Jul 2010 20:22 On Mon, 05 Jul 2010 14:32:13 -0500, Tim Chase wrote: > On 07/05/2010 02:50 AM, Gregor Horvath wrote: >> Am Sun, 04 Jul 2010 18:51:54 -0500 >> schrieb Tim Chase<python.list(a)tim.thechases.com>: >> >>> I think it's the same venting of frustration that caused veteran VB6 >>> developers to start calling VB.Net "Visual Fred" -- the language was >>> too different and too non-backwards-compatible. >>> >>> >> VB6 -> VB.NET and Python 2 -> 3 is not a valid comparison. >> >> VB6 and VB.NET are totally different languages and technologies, with >> some similarity in syntax. This is not true for Python 2->3. This is an >> healthy organic language growth, not an abandon of a language. > > The quintessential example is Py3's breaking of Hello World. It's a > spectrum of language changes -- Visual Fred just happens to be MUCH > further down the same spectrum having more dramatic changes. Only a > subset of $OLD_VER (whether Py2 or VB6) code will run unmodified under > $NEW_VER (whether Py3 or VB.Net). The same holds for older versions of Python to Python 2.5 or 2.6, it's just that you have to work harder to find the incompatibilities. Try running this under later versions: [steve(a)sylar ~]$ python2.0 Python 2.0.1 (#1, Jan 14 2010, 15:43:17) [GCC 4.1.2 20070925 (Red Hat 4.1.2-27)] on linux2 Type "copyright", "credits" or "license" for more information. >>> None = "Hello world" >>> print None Hello world > It just happens that the subset for > Python is considerably larger than the subset for VB (and Python's > conversion tools seem a little more useful than VB's, IMHO). IIRC, > neither raw VB6 nor Py2 byte-code will run raw in the new environment > (old VB .exe files don't make use of .Net libraries/CLR, nor do Py2 .pyc > files run under Py3) Python byte-code has never been compatible across minor version numbers, let alone major ones. -- Steven
From: Gregor Horvath on 6 Jul 2010 06:27
Am Mon, 05 Jul 2010 14:32:13 -0500 schrieb Tim Chase <python.list(a)tim.thechases.com>: > On 07/05/2010 02:50 AM, Gregor Horvath wrote: > > Am Sun, 04 Jul 2010 18:51:54 -0500 > > schrieb Tim Chase<python.list(a)tim.thechases.com>: > > > >> I think it's the same venting of frustration that caused veteran > >> VB6 developers to start calling VB.Net "Visual Fred" -- the > >> language was too different and too non-backwards-compatible. > >> > > > > VB6 -> VB.NET and Python 2 -> 3 is not a valid comparison. > > > > VB6 and VB.NET are totally different languages and technologies, > > with some similarity in syntax. This is not true for Python 2->3. > > This is an healthy organic language growth, not an abandon of a > > language. > > The quintessential example is Py3's breaking of Hello World. > It's a spectrum of language changes -- Visual Fred just happens > to be MUCH further down the same spectrum having more dramatic > changes. Only a subset of $OLD_VER (whether Py2 or VB6) code > will run unmodified under $NEW_VER (whether Py3 or VB.Net). It Don't you think that there is a really huge difference in an evolutionary development of a language with some well founded incompatibilities due to some muck outs on one side and and on the other side stopping the development of a language and replacing it with one derived from a completely different one and giving it a related name and syntax? And that such a big difference forbids any comparison, although there are some superficial similarities? -- Greg |