This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Remove the MANIFEST check from the release guide
[perl5.git] / pod / perlpolicy.pod
CommitLineData
48cb5b3a 1=head1 NAME
3c78fafa 2
9a7064ee 3perlpolicy - Various and sundry policies and commitments related to the Perl core
48cb5b3a
JV
4
5=head1 DESCRIPTION
6
7This document is the master document which records all written
8policies about how the Perl 5 Porters collectively develop and maintain
9the Perl core.
10
a101a770
JV
11=head1 GOVERNANCE
12
13=head2 Perl 5 Porters
14
15Subscribers to perl5-porters (the porters themselves) come in several flavours.
16Some are quiet curious lurkers, who rarely pitch in and instead watch
17the ongoing development to ensure they're forewarned of new changes or
18features in Perl. Some are representatives of vendors, who are there
19to make sure that Perl continues to compile and work on their
20platforms. Some patch any reported bug that they know how to fix,
21some are actively patching their pet area (threads, Win32, the regexp
22-engine), while others seem to do nothing but complain. In other
23words, it's your usual mix of technical people.
24
25Over this group of porters presides Larry Wall. He has the final word
26in what does and does not change in any of the Perl programming languages.
27These days, Larry spends most of his time on Perl 6, while Perl 5 is
28shepherded by a "pumpking", a porter responsible for deciding what
29goes into each release and ensuring that releases happen on a regular
30basis.
31
32Larry sees Perl development along the lines of the US government:
33there's the Legislature (the porters), the Executive branch (the
34-pumpking), and the Supreme Court (Larry). The legislature can
35discuss and submit patches to the executive branch all they like, but
36the executive branch is free to veto them. Rarely, the Supreme Court
37will side with the executive branch over the legislature, or the
38legislature over the executive branch. Mostly, however, the
39legislature and the executive branch are supposed to get along and
40work out their differences without impeachment or court cases.
41
42You might sometimes see reference to Rule 1 and Rule 2. Larry's power
43as Supreme Court is expressed in The Rules:
44
45=over 4
46
47=item 1
48
49Larry is always by definition right about how Perl should behave.
50This means he has final veto power on the core functionality.
51
52=item 2
53
54Larry is allowed to change his mind about any matter at a later date,
55regardless of whether he previously invoked Rule 1.
56
57=back
58
59Got that? Larry is always right, even when he was wrong. It's rare
60to see either Rule exercised, but they are often alluded to.
61
70eadc36
JV
62=head1 MAINTENANCE AND SUPPORT
63
64Perl 5 is developed by a community, not a corporate entity. Every change
65contributed to the Perl core is the result of a donation. Typically, these
66donations are contributions of code or time by individual members of our
67community. On occasion, these donations come in the form of corporate
68or organizational sponsorship of a particular individual or project.
69
70As a volunteer organization, the commitments we make are heavily dependent
71on the goodwill and hard work of individuals who have no obligation to
72contribute to Perl.
73
3b4ebcde 74That being said, we value Perl's stability and security and have long
70eadc36
JV
75had an unwritten covenant with the broader Perl community to support
76and maintain releases of Perl.
77
78This document codifies the support and maintenance commitments that
79the Perl community should expect from Perl's developers:
80
81=over
82
83=item *
84
85We "officially" support the two most recent stable release
e26b5c49 86series. As of the release of 5.14.0, we will "officially"
70eadc36
JV
87end support for Perl 5.10, other than providing security
88updates as described below.
89
90=item *
91
92To the best of our ability, we will attempt to fix critical issues
e26b5c49 93in the two most recent stable 5.x release series. Fixes for the
70eadc36
JV
94current release series take precedence over fixes for the previous
95release series.
96
97=item *
98
99To the best of our ability, we will provide "critical" security patches
f50f542d 100/ releases for any major version of Perl whose 5.x.0 release was within
70a565f4
RS
101the past three years. We can only commit to providing these for the
102most recent .y release in any 5.x.y series.
70eadc36
JV
103
104=item *
105
106We will not provide security updates or bug fixes for development
107releases of Perl.
108
109=item *
110
111We encourage vendors to ship the most recent supported release of
112Perl at the time of their code freeze.
113
114=item *
115
116As a vendor, you may have a requirement to backport security fixes
117beyond our 3 year support commitment. We can provide limited support and
118advice to you as you do so and, where possible will try to apply
119those patches to the relevant -maint branches in git, though we may or
120may not choose to make numbered releases or "official" patches
121available. Contact us at E<lt>perl5-security-report@perl.orgE<gt>
122to begin that process.
123
124=back
125
70e4a83b
JV
126=head1 BACKWARD COMPATIBILITY AND DEPRECATION
127
128Our community has a long-held belief that backward-compatibility is a
129virtue, even when the functionality in question is a design flaw.
130
131We would all love to unmake some mistakes we've made over the past
132decades. Living with every design error we've ever made can lead
133to painful stagnation. Unwinding our mistakes is very, very
134difficult. Doing so without actively harming our users is
135nearly impossible.
136
137Lately, ignoring or actively opposing compatibility with earlier versions
138of Perl has come into vogue. Sometimes, a change is proposed which
139wants to usurp syntax which previously had another meaning. Sometimes,
339a461d 140a change wants to improve previously-crazy semantics.
70e4a83b
JV
141
142Down this road lies madness.
143
144Requiring end-user programmers to change just a few language constructs,
145even language constructs which no well-educated developer would ever
146intentionally use is tantamount to saying "you should not upgrade to
147a new release of Perl unless you have 100% test coverage and can do a
148full manual audit of your codebase." If we were to have tools capable of
149reliably upgrading Perl source code from one version of Perl to another,
150this concern could be significantly mitigated.
151
152We want to ensure that Perl continues to grow and flourish in the coming
153years and decades, but not at the expense of our user community.
154
155Existing syntax and semantics should only be marked for destruction in
156very limited circumstances. If a given language feature's continued
157inclusion in the language will cause significant harm to the language
158or prevent us from making needed changes to the runtime, then it may
159be considered for deprecation.
160
161Any language change which breaks backward-compatibility should be able to
162be enabled or disabled lexically. Unless code at a given scope declares
163that it wants the new behavior, that new behavior should be disabled.
164Which backward-incompatible changes are controlled implicitly by a
165'use v5.x.y' is a decision which should be made by the pumpking in
166consultation with the community.
167
168When a backward-incompatible change can't be toggled lexically, the decision
169to change the language must be considered very, very carefully. If it's
170possible to move the old syntax or semantics out of the core language
171and into XS-land, that XS module should be enabled by default unless
172the user declares that they want a newer revision of Perl.
173
174Historically, we've held ourselves to a far higher standard than
175backward-compatibility -- bugward-compatibility. Any accident of
176implementation or unintentional side-effect of running some bit of code
177has been considered to be a feature of the language to be defended with
178the same zeal as any other feature or functionality. No matter how
179frustrating these unintentional features may be to us as we continue
180to improve Perl, these unintentional features often deserve our
181protection. It is very important that existing software written in
182Perl continue to work correctly. If end-user developers have adopted a
183bug as a feature, we need to treat it as such.
184
185New syntax and semantics which don't break existing language constructs
186and syntax have a much lower bar. They merely need to prove themselves
b6538e4f 187to be useful, elegant, well designed, and well tested.
70e4a83b
JV
188
189=head2 Terminology
190
191To make sure we're talking about the same thing when we discuss the removal
192of features or functionality from the Perl core, we have specific definitions
193for a few words and phrases.
194
195=over
196
197=item experimental
198
199If something in the Perl core is marked as B<experimental>, we may change
200its behaviour, deprecate or remove it without notice. While we'll always
201do our best to smooth the transition path for users of experimental
202features, you should contact the perl5-porters mailinglist if you find
203an experimental feature useful and want to help shape its future.
204
205=item deprecated
206
207If something in the Perl core is marked as B<deprecated>, we may remove it
3b4ebcde 208from the core in the next stable release series, though we may not. As of
70e4a83b
JV
209Perl 5.12, deprecated features and modules warn the user as they're used.
210If you use a deprecated feature and believe that its removal from the Perl
211core would be a mistake, please contact the perl5-porters mailinglist and
212plead your case. We don't deprecate things without a good reason, but
213sometimes there's a counterargument we haven't considered. Historically,
214we did not distinguish between "deprecated" and "discouraged" features.
215
216=item discouraged
217
218From time to time, we may mark language constructs and features which we
219consider to have been mistakes as B<discouraged>. Discouraged features
220aren't candidates for removal in the next major release series, but
221we may later deprecate them if they're found to stand in the way of a
9a7064ee 222significant improvement to the Perl core.
70e4a83b
JV
223
224=item removed
225
226Once a feature, construct or module has been marked as deprecated for a
9a7064ee 227stable release cycle, we may remove it from the Perl core. Unsurprisingly,
70e4a83b
JV
228we say we've B<removed> these things.
229
230=back
48cb5b3a 231
fcf56c88
JV
232=head1 MAINTENANCE BRANCHES
233
234=over
235
236=item *
237
238New releases of maint should contain as few changes as possible.
239If there is any question about whether a given patch might merit
240inclusion in a maint release, then it almost certainly should not
241be included.
242
243=item *
244
245Portability fixes, such as changes to Configure and the files in
246hints/ are acceptable. Ports of Perl to a new platform, architecture
247or OS release that involve changes to the implementation are NOT
248acceptable.
249
250=item *
251
b6538e4f
TC
252Acceptable documentation updates are those that correct factual errors,
253explain significant bugs or deficiencies in the current implementation,
254or fix broken markup.
fcf56c88
JV
255
256=item *
257
258Patches that add new warnings or errors or deprecate features
259are not acceptable.
260
261=item *
262
263Patches that fix crashing bugs that do not otherwise change Perl's
264functionality or negatively impact performance are acceptable.
265
266=item *
267
268Patches that fix CVEs or security issues are acceptable, but should
269be run through the perl5-security-report@perl.org mailing list
270rather than applied directly.
271
272=item *
273
274Updates to dual-life modules should consist of minimal patches to
275fix crashing or security issues (as above).
276
277=item *
278
27d0393b
JV
279Minimal patches that fix platform-specific test failures or
280installation issues are acceptable. When these changes are made
281to dual-life modules for which CPAN is canonical, any changes
282should be coordinated with the upstream author.
283
284=item *
285
fcf56c88
JV
286New versions of dual-life modules should NOT be imported into maint.
287Those belong in the next stable series.
288
289=item *
290
291Patches that add or remove features are not acceptable.
292
293=item *
294
295Patches that break binary compatibility are not acceptable. (Please
296talk to a pumpking.)
297
298=back
299
300
301=head2 Getting changes into a maint branch
302
303Historically, only the pumpking cherry-picked changes from bleadperl
304into maintperl. This has...scaling problems. At the same time,
305maintenance branches of stable versions of Perl need to be treated with
306great care. To that end, we're going to try out a new process for
307maint-5.12.
308
309Any committer may cherry-pick any commit from blead to maint-5.12 if
310they send mail to perl5-porters announcing their intent to cherry-pick
311a specific commit along with a rationale for doing so and at least two
312other committers respond to the list giving their assent. (This policy
313applies to current and former pumpkings, as well as other committers.)
48cb5b3a
JV
314
315=head1 CONTRIBUTED MODULES
316
317
318=head2 A Social Contract about Artistic Control
6ee623d5
GS
319
320What follows is a statement about artistic control, defined as the ability
321of authors of packages to guide the future of their code and maintain
322control over their work. It is a recognition that authors should have
323control over their work, and that it is a responsibility of the rest of
324the Perl community to ensure that they retain this control. It is an
325attempt to document the standards to which we, as Perl developers, intend
326to hold ourselves. It is an attempt to write down rough guidelines about
327the respect we owe each other as Perl developers.
328
329This statement is not a legal contract. This statement is not a legal
330document in any way, shape, or form. Perl is distributed under the GNU
331Public License and under the Artistic License; those are the precise legal
332terms. This statement isn't about the law or licenses. It's about
333community, mutual respect, trust, and good-faith cooperation.
334
335We recognize that the Perl core, defined as the software distributed with
336the heart of Perl itself, is a joint project on the part of all of us.
aaa2bbb1 337From time to time, a script, module, or set of modules (hereafter referred
6ee623d5
GS
338to simply as a "module") will prove so widely useful and/or so integral to
339the correct functioning of Perl itself that it should be distributed with
9a7064ee 340the Perl core. This should never be done without the author's explicit
6ee623d5
GS
341consent, and a clear recognition on all parts that this means the module
342is being distributed under the same terms as Perl itself. A module author
343should realize that inclusion of a module into the Perl core will
344necessarily mean some loss of control over it, since changes may
345occasionally have to be made on short notice or for consistency with the
346rest of Perl.
347
348Once a module has been included in the Perl core, however, everyone
349involved in maintaining Perl should be aware that the module is still the
350property of the original author unless the original author explicitly
351gives up their ownership of it. In particular:
352
48cb5b3a
JV
353=over
354
171407a0
JJ
355=item *
356
9a7064ee 357The version of the module in the Perl core should still be considered the
171407a0
JJ
358work of the original author. All patches, bug reports, and so
359forth should be fed back to them. Their development directions
360should be respected whenever possible.
6ee623d5 361
48cb5b3a
JV
362=item *
363
364Patches may be applied by the pumpkin holder without the explicit
365cooperation of the module author if and only if they are very minor,
366time-critical in some fashion (such as urgent security fixes), or if
367the module author cannot be reached. Those patches must still be
368given back to the author when possible, and if the author decides on
369an alternate fix in their version, that fix should be strongly
370preferred unless there is a serious problem with it. Any changes not
371endorsed by the author should be marked as such, and the contributor
372of the change acknowledged.
373
374=item *
375
376The version of the module distributed with Perl should, whenever
377possible, be the latest version of the module as distributed by the
378author (the latest non-beta version in the case of public Perl
379releases), although the pumpkin holder may hold off on upgrading the
380version of the module distributed with Perl to the latest version
381until the latest version has had sufficient testing.
382
383=back
6ee623d5
GS
384
385In other words, the author of a module should be considered to have final
386say on modifications to their module whenever possible (bearing in mind
387that it's expected that everyone involved will work together and arrive at
388reasonable compromises when there are disagreements).
389
390As a last resort, however:
391
48cb5b3a
JV
392
393If the author's vision of the future of their module is sufficiently
394different from the vision of the pumpkin holder and perl5-porters as a
395whole so as to cause serious problems for Perl, the pumpkin holder may
9a7064ee 396choose to formally fork the version of the module in the Perl core from the
48cb5b3a 397one maintained by the author. This should not be done lightly and
c4f5d98d 398should B<always> if at all possible be done only after direct input
48cb5b3a 399from Larry. If this is done, it must then be made explicit in the
9a7064ee 400module as distributed with the Perl core that it is a forked version and
48cb5b3a
JV
401that while it is based on the original author's work, it is no longer
402maintained by them. This must be noted in both the documentation and
403in the comments in the source of the module.
6ee623d5
GS
404
405Again, this should be a last resort only. Ideally, this should never
406happen, and every possible effort at cooperation and compromise should be
407made before doing this. If it does prove necessary to fork a module for
408the overall health of Perl, proper credit must be given to the original
409author in perpetuity and the decision should be constantly re-evaluated to
410see if a remerging of the two branches is possible down the road.
411
412In all dealings with contributed modules, everyone maintaining Perl should
413keep in mind that the code belongs to the original author, that they may
414not be on perl5-porters at any given time, and that a patch is not
415official unless it has been integrated into the author's copy of the
416module. To aid with this, and with points #1, #2, and #3 above, contact
417information for the authors of all contributed modules should be kept with
418the Perl distribution.
419
420Finally, the Perl community as a whole recognizes that respect for
421ownership of code, respect for artistic control, proper credit, and active
422effort to prevent unintentional code skew or communication gaps is vital
423to the health of the community and Perl itself. Members of a community
424should not normally have to resort to rules and laws to deal with each
425other, and this document, although it contains rules so as to be clear, is
426about an attitude and general approach. The first step in any dispute
427should be open communication, respect for opposing views, and an attempt
428at a compromise. In nearly every circumstance nothing more will be
429necessary, and certainly no more drastic measure should be used until
430every avenue of communication and discussion has failed.
3c78fafa 431
70e4a83b 432
3b4ebcde
JV
433=head1 DOCUMENTATION
434
435Perl's documentation is an important resource for our users. It's
436incredibly important for Perl's documentation to be reasonably coherent
437and to accurately reflect the current implementation.
438
439Just as P5P collectively maintains the codebase, we collectively
440maintain the documentation. Writing a particular bit of documentation
441doesn't give an author control of the future of that documentation.
442At the same time, just as source code changes should match the style
443of their surrounding blocks, so should documentation changes.
444
445Examples in documentation should be illustrative of the concept
446they're explaining. Sometimes, the best way to show how a
447language feature works is with a small program the reader can
448run without modification. More often, examples will consist
449of a snippet of code containing only the "important" bits.
450The definition of "important" varies from snippet to snippet.
1bb8a155 451Sometimes it's important to declare C<use strict> and C<use warnings>,
3b4ebcde
JV
452initialize all variables and fully catch every error condition.
453More often than not, though, those things obscure the lesson
454the example was intended to teach.
455
456As Perl is developed by a global team of volunteers, our
457documentation often contains spellings which look funny
458to I<somebody>. Choice of American/British/Other spellings
459is left as an exercise for the author of each bit of
460documentation. When patching documentation, try to emulate
461the documentation around you, rather than changing the existing
462prose.
463
464In general, documentation should describe what Perl does "now" rather
465than what it used to do. It's perfectly reasonable to include notes
466in documentation about how behaviour has changed from previous releases,
9e9fdd5d 467but, with very few exceptions, documentation isn't "dual-life" --
3b4ebcde
JV
468it doesn't need to fully describe how all old versions used to work.
469
470
48cb5b3a
JV
471=head1 CREDITS
472
3b4ebcde 473"Social Contract about Contributed Modules" originally by Russ Allbery E<lt>rra@stanford.eduE<gt> and the perl5-porters.
3c78fafa 474