1=encoding utf8 2 3=head1 NAME 4 5perlpolicy - Various and sundry policies and commitments related to the Perl core 6 7=head1 DESCRIPTION 8 9This document is the master document which records all written 10policies about how the Perl 5 Porters collectively develop and maintain 11the Perl core. 12 13=head1 GOVERNANCE 14 15=head2 Perl 5 Porters 16 17Subscribers to perl5-porters (the porters themselves) come in several flavours. 18Some are quiet curious lurkers, who rarely pitch in and instead watch 19the ongoing development to ensure they're forewarned of new changes or 20features in Perl. Some are representatives of vendors, who are there 21to make sure that Perl continues to compile and work on their 22platforms. Some patch any reported bug that they know how to fix, 23some are actively patching their pet area (threads, Win32, the regexp 24-engine), while others seem to do nothing but complain. In other 25words, it's your usual mix of technical people. 26 27Over this group of porters presides Larry Wall. He has the final word 28in what does and does not change in any of the Perl programming languages. 29These days, Larry spends most of his time on Perl 6, while Perl 5 is 30shepherded by a "pumpking", a porter responsible for deciding what 31goes into each release and ensuring that releases happen on a regular 32basis. 33 34Larry sees Perl development along the lines of the US government: 35there's the Legislature (the porters), the Executive branch (the 36-pumpking), and the Supreme Court (Larry). The legislature can 37discuss and submit patches to the executive branch all they like, but 38the executive branch is free to veto them. Rarely, the Supreme Court 39will side with the executive branch over the legislature, or the 40legislature over the executive branch. Mostly, however, the 41legislature and the executive branch are supposed to get along and 42work out their differences without impeachment or court cases. 43 44You might sometimes see reference to Rule 1 and Rule 2. Larry's power 45as Supreme Court is expressed in The Rules: 46 47=over 4 48 49=item 1 50 51Larry is always by definition right about how Perl should behave. 52This means he has final veto power on the core functionality. 53 54=item 2 55 56Larry is allowed to change his mind about any matter at a later date, 57regardless of whether he previously invoked Rule 1. 58 59=back 60 61Got that? Larry is always right, even when he was wrong. It's rare 62to see either Rule exercised, but they are often alluded to. 63 64=head1 MAINTENANCE AND SUPPORT 65 66Perl 5 is developed by a community, not a corporate entity. Every change 67contributed to the Perl core is the result of a donation. Typically, these 68donations are contributions of code or time by individual members of our 69community. On occasion, these donations come in the form of corporate 70or organizational sponsorship of a particular individual or project. 71 72As a volunteer organization, the commitments we make are heavily dependent 73on the goodwill and hard work of individuals who have no obligation to 74contribute to Perl. 75 76That being said, we value Perl's stability and security and have long 77had an unwritten covenant with the broader Perl community to support 78and maintain releases of Perl. 79 80This document codifies the support and maintenance commitments that 81the Perl community should expect from Perl's developers: 82 83=over 84 85=item * 86 87We "officially" support the two most recent stable release series. 5.16.x 88and earlier are now out of support. As of the release of 5.22.0, we will 89"officially" end support for Perl 5.18.x, other than providing security 90updates as described below. 91 92=item * 93 94To the best of our ability, we will attempt to fix critical issues 95in the two most recent stable 5.x release series. Fixes for the 96current release series take precedence over fixes for the previous 97release series. 98 99=item * 100 101To the best of our ability, we will provide "critical" security patches 102/ releases for any major version of Perl whose 5.x.0 release was within 103the past three years. We can only commit to providing these for the 104most recent .y release in any 5.x.y series. 105 106=item * 107 108We will not provide security updates or bug fixes for development 109releases of Perl. 110 111=item * 112 113We encourage vendors to ship the most recent supported release of 114Perl at the time of their code freeze. 115 116=item * 117 118As a vendor, you may have a requirement to backport security fixes 119beyond our 3 year support commitment. We can provide limited support and 120advice to you as you do so and, where possible will try to apply 121those patches to the relevant -maint branches in git, though we may or 122may not choose to make numbered releases or "official" patches 123available. Contact us at E<lt>perl5-security-report@perl.orgE<gt> 124to begin that process. 125 126=back 127 128=head1 BACKWARD COMPATIBILITY AND DEPRECATION 129 130Our community has a long-held belief that backward-compatibility is a 131virtue, even when the functionality in question is a design flaw. 132 133We would all love to unmake some mistakes we've made over the past 134decades. Living with every design error we've ever made can lead 135to painful stagnation. Unwinding our mistakes is very, very 136difficult. Doing so without actively harming our users is 137nearly impossible. 138 139Lately, ignoring or actively opposing compatibility with earlier versions 140of Perl has come into vogue. Sometimes, a change is proposed which 141wants to usurp syntax which previously had another meaning. Sometimes, 142a change wants to improve previously-crazy semantics. 143 144Down this road lies madness. 145 146Requiring end-user programmers to change just a few language constructs, 147even language constructs which no well-educated developer would ever 148intentionally use is tantamount to saying "you should not upgrade to 149a new release of Perl unless you have 100% test coverage and can do a 150full manual audit of your codebase." If we were to have tools capable of 151reliably upgrading Perl source code from one version of Perl to another, 152this concern could be significantly mitigated. 153 154We want to ensure that Perl continues to grow and flourish in the coming 155years and decades, but not at the expense of our user community. 156 157Existing syntax and semantics should only be marked for destruction in 158very limited circumstances. If a given language feature's continued 159inclusion in the language will cause significant harm to the language 160or prevent us from making needed changes to the runtime, then it may 161be considered for deprecation. 162 163Any language change which breaks backward-compatibility should be able to 164be enabled or disabled lexically. Unless code at a given scope declares 165that it wants the new behavior, that new behavior should be disabled. 166Which backward-incompatible changes are controlled implicitly by a 167'use v5.x.y' is a decision which should be made by the pumpking in 168consultation with the community. 169 170When a backward-incompatible change can't be toggled lexically, the decision 171to change the language must be considered very, very carefully. If it's 172possible to move the old syntax or semantics out of the core language 173and into XS-land, that XS module should be enabled by default unless 174the user declares that they want a newer revision of Perl. 175 176Historically, we've held ourselves to a far higher standard than 177backward-compatibility -- bugward-compatibility. Any accident of 178implementation or unintentional side-effect of running some bit of code 179has been considered to be a feature of the language to be defended with 180the same zeal as any other feature or functionality. No matter how 181frustrating these unintentional features may be to us as we continue 182to improve Perl, these unintentional features often deserve our 183protection. It is very important that existing software written in 184Perl continue to work correctly. If end-user developers have adopted a 185bug as a feature, we need to treat it as such. 186 187New syntax and semantics which don't break existing language constructs 188and syntax have a much lower bar. They merely need to prove themselves 189to be useful, elegant, well designed, and well tested. 190 191=head2 Terminology 192 193To make sure we're talking about the same thing when we discuss the removal 194of features or functionality from the Perl core, we have specific definitions 195for a few words and phrases. 196 197=over 198 199=item experimental 200 201If something in the Perl core is marked as B<experimental>, we may change 202its behaviour, deprecate or remove it without notice. While we'll always 203do our best to smooth the transition path for users of experimental 204features, you should contact the perl5-porters mailinglist if you find 205an experimental feature useful and want to help shape its future. 206 207Experimental features must be experimental in two stable releases before being 208marked non-experimental. Experimental features will only have their 209experimental status revoked when they no longer have any design-changing bugs 210open against them and when they have remained unchanged in behavior for the 211entire length of a development cycle. In other words, a feature present in 212v5.20.0 may be marked no longer experimental in v5.22.0 if and only if its 213behavior is unchanged throughout all of v5.21. 214 215=item deprecated 216 217If something in the Perl core is marked as B<deprecated>, we may remove it 218from the core in the future, though we might not. Generally, backward 219incompatible changes will have deprecation warnings for two release 220cycles before being removed, but may be removed after just one cycle if 221the risk seems quite low or the benefits quite high. 222 223As of 224Perl 5.12, deprecated features and modules warn the user as they're used. 225When a module is deprecated, it will also be made available on CPAN. 226Installing it from CPAN will silence deprecation warnings for that module. 227 228If you use a deprecated feature or module and believe that its removal from 229the Perl core would be a mistake, please contact the perl5-porters 230mailinglist and plead your case. We don't deprecate things without a good 231reason, but sometimes there's a counterargument we haven't considered. 232Historically, we did not distinguish between "deprecated" and "discouraged" 233features. 234 235=item discouraged 236 237From time to time, we may mark language constructs and features which we 238consider to have been mistakes as B<discouraged>. Discouraged features 239aren't currently candidates for removal, but 240we may later deprecate them if they're found to stand in the way of a 241significant improvement to the Perl core. 242 243=item removed 244 245Once a feature, construct or module has been marked as deprecated, we 246may remove it from the Perl core. Unsurprisingly, 247we say we've B<removed> these things. When a module is removed, it will 248no longer ship with Perl, but will continue to be available on CPAN. 249 250=back 251 252=head1 MAINTENANCE BRANCHES 253 254=over 255 256=item * 257 258New releases of maint should contain as few changes as possible. 259If there is any question about whether a given patch might merit 260inclusion in a maint release, then it almost certainly should not 261be included. 262 263=item * 264 265Portability fixes, such as changes to Configure and the files in 266hints/ are acceptable. Ports of Perl to a new platform, architecture 267or OS release that involve changes to the implementation are NOT 268acceptable. 269 270=item * 271 272Acceptable documentation updates are those that correct factual errors, 273explain significant bugs or deficiencies in the current implementation, 274or fix broken markup. 275 276=item * 277 278Patches that add new warnings or errors or deprecate features 279are not acceptable. 280 281=item * 282 283Patches that fix crashing bugs, assertion failures and 284memory corruption that do not otherwise change Perl's 285functionality or negatively impact performance are acceptable. 286 287=item * 288 289Patches that fix CVEs or security issues are acceptable, but should 290be run through the perl5-security-report@perl.org mailing list 291rather than applied directly. 292 293=item * 294 295Patches that fix regressions in perl's behavior relative to previous 296releases are acceptable. 297 298=item * 299 300Updates to dual-life modules should consist of minimal patches to 301fix crashing or security issues (as above). 302 303=item * 304 305Minimal patches that fix platform-specific test failures or build or 306installation issues are acceptable. When these changes are made 307to dual-life modules for which CPAN is canonical, any changes 308should be coordinated with the upstream author. 309 310=item * 311 312New versions of dual-life modules should NOT be imported into maint. 313Those belong in the next stable series. 314 315=item * 316 317Patches that add or remove features are not acceptable. 318 319=item * 320 321Patches that break binary compatibility are not acceptable. (Please 322talk to a pumpking.) 323 324=back 325 326 327=head2 Getting changes into a maint branch 328 329Historically, only the pumpking cherry-picked changes from bleadperl 330into maintperl. This has scaling problems. At the same time, 331maintenance branches of stable versions of Perl need to be treated with 332great care. To that end, as of Perl 5.12, we have a new process for 333maint branches. 334 335Any committer may cherry-pick any commit from blead to a maint branch if 336they send mail to perl5-porters announcing their intent to cherry-pick 337a specific commit along with a rationale for doing so and at least two 338other committers respond to the list giving their assent. (This policy 339applies to current and former pumpkings, as well as other committers.) 340 341=head1 CONTRIBUTED MODULES 342 343 344=head2 A Social Contract about Artistic Control 345 346What follows is a statement about artistic control, defined as the ability 347of authors of packages to guide the future of their code and maintain 348control over their work. It is a recognition that authors should have 349control over their work, and that it is a responsibility of the rest of 350the Perl community to ensure that they retain this control. It is an 351attempt to document the standards to which we, as Perl developers, intend 352to hold ourselves. It is an attempt to write down rough guidelines about 353the respect we owe each other as Perl developers. 354 355This statement is not a legal contract. This statement is not a legal 356document in any way, shape, or form. Perl is distributed under the GNU 357Public License and under the Artistic License; those are the precise legal 358terms. This statement isn't about the law or licenses. It's about 359community, mutual respect, trust, and good-faith cooperation. 360 361We recognize that the Perl core, defined as the software distributed with 362the heart of Perl itself, is a joint project on the part of all of us. 363From time to time, a script, module, or set of modules (hereafter referred 364to simply as a "module") will prove so widely useful and/or so integral to 365the correct functioning of Perl itself that it should be distributed with 366the Perl core. This should never be done without the author's explicit 367consent, and a clear recognition on all parts that this means the module 368is being distributed under the same terms as Perl itself. A module author 369should realize that inclusion of a module into the Perl core will 370necessarily mean some loss of control over it, since changes may 371occasionally have to be made on short notice or for consistency with the 372rest of Perl. 373 374Once a module has been included in the Perl core, however, everyone 375involved in maintaining Perl should be aware that the module is still the 376property of the original author unless the original author explicitly 377gives up their ownership of it. In particular: 378 379=over 380 381=item * 382 383The version of the module in the Perl core should still be considered the 384work of the original author. All patches, bug reports, and so 385forth should be fed back to them. Their development directions 386should be respected whenever possible. 387 388=item * 389 390Patches may be applied by the pumpkin holder without the explicit 391cooperation of the module author if and only if they are very minor, 392time-critical in some fashion (such as urgent security fixes), or if 393the module author cannot be reached. Those patches must still be 394given back to the author when possible, and if the author decides on 395an alternate fix in their version, that fix should be strongly 396preferred unless there is a serious problem with it. Any changes not 397endorsed by the author should be marked as such, and the contributor 398of the change acknowledged. 399 400=item * 401 402The version of the module distributed with Perl should, whenever 403possible, be the latest version of the module as distributed by the 404author (the latest non-beta version in the case of public Perl 405releases), although the pumpkin holder may hold off on upgrading the 406version of the module distributed with Perl to the latest version 407until the latest version has had sufficient testing. 408 409=back 410 411In other words, the author of a module should be considered to have final 412say on modifications to their module whenever possible (bearing in mind 413that it's expected that everyone involved will work together and arrive at 414reasonable compromises when there are disagreements). 415 416As a last resort, however: 417 418 419If the author's vision of the future of their module is sufficiently 420different from the vision of the pumpkin holder and perl5-porters as a 421whole so as to cause serious problems for Perl, the pumpkin holder may 422choose to formally fork the version of the module in the Perl core from the 423one maintained by the author. This should not be done lightly and 424should B<always> if at all possible be done only after direct input 425from Larry. If this is done, it must then be made explicit in the 426module as distributed with the Perl core that it is a forked version and 427that while it is based on the original author's work, it is no longer 428maintained by them. This must be noted in both the documentation and 429in the comments in the source of the module. 430 431Again, this should be a last resort only. Ideally, this should never 432happen, and every possible effort at cooperation and compromise should be 433made before doing this. If it does prove necessary to fork a module for 434the overall health of Perl, proper credit must be given to the original 435author in perpetuity and the decision should be constantly re-evaluated to 436see if a remerging of the two branches is possible down the road. 437 438In all dealings with contributed modules, everyone maintaining Perl should 439keep in mind that the code belongs to the original author, that they may 440not be on perl5-porters at any given time, and that a patch is not 441official unless it has been integrated into the author's copy of the 442module. To aid with this, and with points #1, #2, and #3 above, contact 443information for the authors of all contributed modules should be kept with 444the Perl distribution. 445 446Finally, the Perl community as a whole recognizes that respect for 447ownership of code, respect for artistic control, proper credit, and active 448effort to prevent unintentional code skew or communication gaps is vital 449to the health of the community and Perl itself. Members of a community 450should not normally have to resort to rules and laws to deal with each 451other, and this document, although it contains rules so as to be clear, is 452about an attitude and general approach. The first step in any dispute 453should be open communication, respect for opposing views, and an attempt 454at a compromise. In nearly every circumstance nothing more will be 455necessary, and certainly no more drastic measure should be used until 456every avenue of communication and discussion has failed. 457 458 459=head1 DOCUMENTATION 460 461Perl's documentation is an important resource for our users. It's 462incredibly important for Perl's documentation to be reasonably coherent 463and to accurately reflect the current implementation. 464 465Just as P5P collectively maintains the codebase, we collectively 466maintain the documentation. Writing a particular bit of documentation 467doesn't give an author control of the future of that documentation. 468At the same time, just as source code changes should match the style 469of their surrounding blocks, so should documentation changes. 470 471Examples in documentation should be illustrative of the concept 472they're explaining. Sometimes, the best way to show how a 473language feature works is with a small program the reader can 474run without modification. More often, examples will consist 475of a snippet of code containing only the "important" bits. 476The definition of "important" varies from snippet to snippet. 477Sometimes it's important to declare C<use strict> and C<use warnings>, 478initialize all variables and fully catch every error condition. 479More often than not, though, those things obscure the lesson 480the example was intended to teach. 481 482As Perl is developed by a global team of volunteers, our 483documentation often contains spellings which look funny 484to I<somebody>. Choice of American/British/Other spellings 485is left as an exercise for the author of each bit of 486documentation. When patching documentation, try to emulate 487the documentation around you, rather than changing the existing 488prose. 489 490In general, documentation should describe what Perl does "now" rather 491than what it used to do. It's perfectly reasonable to include notes 492in documentation about how behaviour has changed from previous releases, 493but, with very few exceptions, documentation isn't "dual-life" -- 494it doesn't need to fully describe how all old versions used to work. 495 496=head1 STANDARDS OF CONDUCT 497 498The official forum for the development of perl is the perl5-porters mailing 499list, mentioned above, and its bugtracker at rt.perl.org. All participants in 500discussion there are expected to adhere to a standard of conduct. 501 502=over 4 503 504=item * 505 506Always be civil. 507 508=item * 509 510Heed the moderators. 511 512=back 513 514Civility is simple: stick to the facts while avoiding demeaning remarks and 515sarcasm. It is not enough to be factual. You must also be civil. Responding 516in kind to incivility is not acceptable. 517 518While civility is required, kindness is encouraged; if you have any doubt about 519whether you are being civil, simply ask yourself, "Am I being kind?" and aspire 520to that. 521 522If the list moderators tell you that you are not being civil, carefully 523consider how your words have appeared before responding in any way. Were they 524kind? You may protest, but repeated protest in the face of a repeatedly 525reaffirmed decision is not acceptable. 526 527Unacceptable behavior will result in a public and clearly identified warning. 528Repeated unacceptable behavior will result in removal from the mailing list and 529revocation of rights to update rt.perl.org. The first removal is for one 530month. Subsequent removals will double in length. After six months with no 531warning, a user's ban length is reset. Removals, like warnings, are public. 532 533The list of moderators will be public knowledge. At present, it is: 534Aaron Crane, Andy Dougherty, Ricardo Signes, Steffen Müller. 535 536=head1 CREDITS 537 538"Social Contract about Contributed Modules" originally by Russ Allbery E<lt>rra@stanford.eduE<gt> and the perl5-porters. 539 540