diff options
Diffstat (limited to 'docs/html/future.html')
| -rw-r--r-- | docs/html/future.html | 609 |
1 files changed, 599 insertions, 10 deletions
diff --git a/docs/html/future.html b/docs/html/future.html index 314f0e050..99a577ff8 100644 --- a/docs/html/future.html +++ b/docs/html/future.html @@ -13,8 +13,8 @@ REL="PREVIOUS" TITLE="Tinderbox/Tinderbox2" HREF="tinderbox.html"><LINK REL="NEXT" -TITLE="The Bugzilla FAQ" -HREF="faq.html"></HEAD +TITLE="Bugzilla Variants and Competitors" +HREF="variants.html"></HEAD ><BODY CLASS="CHAPTER" BGCOLOR="#FFFFFF" @@ -54,7 +54,7 @@ WIDTH="10%" ALIGN="right" VALIGN="bottom" ><A -HREF="faq.html" +HREF="variants.html" >Next</A ></TD ></TR @@ -85,11 +85,600 @@ CLASS="SYNOPSIS" ></TR ></TABLE ><P -> The future of Bugzilla is Bugzilla 3.0. Unfortunately, I do - not have more information about it right now, and most of what - went into the "future" section is now present. That stuff was - blue-sky a year ago; MattyT should have me a new document - sometime... +> Bugzilla's future is a constantly-changing thing, as various developers + <SPAN +CLASS="QUOTE" +>"scratch an itch"</SPAN +> when it comes to functionality. + Thus this section is very malleable, subject to change without notice, etc. + You'll probably also notice the lack of formatting. I apologize that it's + not quite as readable as the rest of the Guide. + </P +><P +> <P +CLASS="LITERALLAYOUT" +> Bugzilla Blue Sky<br> +<br> +Customisability<br> +<br> + One of the major stumbling blocks of Bugzilla has been that it is too<br> + rigid and does not adapt itself well enough to the needs of an<br> + organisation. This has led to organisations making changes to the<br> + Bugzilla code that need to be redone each new version of Bugzilla.<br> + Bugzilla should attempt to move away from this to a world where this<br> + doesn't need to occur.<br> +<br> + Most of the subsections in this section are currently explicit design<br> + goals for the "Bugzilla 3" rewrite. This does not necessarily mean<br> + that they will not occur before them in Bugzilla 2, but most are<br> + significant undertakings.<br> +<br> + Field Customisation<br> +<br> + Many installations wish to customise the fields that appear on bug<br> + reports. Current versions of Bugzilla offer limited<br> + customisability. In particular, some fields can be turned off.<br> +<br> + However, many administrators wish to add their own fields, and rename<br> + or otherwise modify existing fields. An architecture that supports<br> + this would be extraordinarily useful.<br> +<br> + Indeed, many fields work similarly and could be abstracted into "field<br> + types", so that an administrator need write little or no code to<br> + support the new fields they desire.<br> +<br> + Possible field types include text (eg status whiteboard), numbers,<br> + dates (eg report time), accounts (eg reporter, qa, cc), inter-bug<br> + relationships (dependencies, duplicates), option groups (platform, os,<br> + severity, priority, target milestone, version) etc.<br> +<br> + Ideally an administrator could configure their fields through a<br> + Bugzilla interface that requires no code to be added. However, it is<br> + highly unlikely this ideal will never be met, and in a similar way<br> + that office applications have scripting languages, Bugzilla should<br> + allow new field types to be written.<br> +<br> + Similarly, a common desire is for resolutions to be added or removed.<br> +<br> + Allocations<br> +<br> + ?<br> +<br> + Option Groups<br> +<br> + ?<br> +<br> + Relations<br> +<br> + ?<br> +<br> + Database Integrity<br> +<br> + Furthermore, it is desirable for administrators to be able to specify<br> + rules that must or should apply between the fields on a bug report.<br> +<br> + For example, you might wish to specify that a bug with status ASSIGNED<br> + must have a target milestone field that that is not untargetted. Or<br> + that a bug with a certain number of votes should get ASSIGNED. Or<br> + that the QA contact must be different from the assignee.<br> +<br> + "Must" relationships could be implemented by refusing to make changes<br> + that violate the relationships, or alternatively, automatically<br> + updating certain fields in order to satisfy the criteria. Which<br> + occurs should be up to the administrator.<br> +<br> + "Should" relationships could be implemented by a combination of<br> + emitting warnings on the process bug page, the same on notification<br> + mails, or emitting periodic whine mails about the situation. Again,<br> + which occurs should be up to the administrator.<br> +<br> + It should also be possible for whine mails to be emitted for "must"<br> + relationships, as they might become violated through direct database<br> + access, Bugzilla bugs, or because they were there before the<br> + relationship was enforced.<br> +<br> + As well as implementing intra-bug constraints, it would be useful to<br> + create inter-bug constraints. For example, a bug that is dependent on<br> + another bug should not have an earlier milestone or greater priority<br> + than that bug.<br> +<br> + Database Adaptability<br> +<br> + Often an administrator desires that fields adapt to the values of<br> + other fields. For example, the value of a field might determine the<br> + possible values of another field or even whether it appears (whether<br> + it is "applicable").<br> +<br> + Limited adaptability is present in Bugzilla 2, and only on the<br> + "Product" field:<br> + * The possible values of the target milestone, version and component<br> + fields depend on the product.<br> + * UNCONFIRMED can be turned off for specific products.<br> + * Voting can be configured differently or turned off for different<br> + products, and there is a separate user vote limits for each<br> + product.<br> +<br> + It would be good if more adaptability was present, both in terms of<br> + all fields relying on the product, as well as the ability to adapt<br> + based on the value of all fields.<br> +<br> + Example ???<br> +<br> + General adaptability raises the issue of circular references between<br> + fields causing problems. One possible solution to this is to place<br> + the fields in a total ordering and require a field refer only to the<br> + previous fields.<br> +<br> + In Bugzilla 2, changing the product of a bug meant a second page would<br> + appear that allowed you to choose a new milestone, component and<br> + version, as those fields adapted themselves to the new product. This<br> + page could be generalised to support all instances where:<br> + * a field value must or might be changed because the possible values<br> + have changed<br> + * is going to drop off because it it is no longer applicable, and<br> + this should be confirmed<br> + * must be specified because it is suddenly applicable, and the<br> + default value, if one exists, might not be acceptable<br> +<br> + Database Independence<br> +<br> + Currently Bugzilla only runs on the MySQL database. It would be<br> + desirable for Bugzilla to run on other databases, because:<br> + * Organisations may have existing database products they use and<br> + would prefer to run a homogenous environment.<br> + * Databases each have their own shortcomings, including MySQL. An<br> + administrator might choose a database that would work better with<br> + their Bugzilla.<br> +<br> + This raises the possibility that we could use features that are only<br> + present in some databases, by appropriately falling back. For<br> + example, in the MySQL world, we live without:<br> + * record-level locking, instead we use table-level locking<br> + * referential and record constraints, instead we checking code<br> + * subselects, instead we use multiple queries and redundant "caches"<br> +<br> + Multiple Front Ends<br> +<br> + Currently Bugzilla is manipulated via the Web, and notifies via<br> + E-Mail. It would be desirable for Bugzilla to easily support various<br> + front ends.<br> +<br> + There is no reason that Bugzilla could not be controlled via a whole<br> + range of front ends, including Web, E-Mail, IRC, ICQ, etc, and<br> + similarly for how it notifies. It's also possible that we could<br> + introduce a special Bugzilla client that uses its own protocol, for<br> + maximum user productivity.<br> +<br> + Indeed a request reply might be returned via a totally different<br> + transport method than was use to submit the request.<br> +<br> +Internationalisation<br> +<br> + Bugzilla currently supports only English. All of the field names,<br> + user instructions, etc are written in English. It would be desirable<br> + to allow "language packs" so Bugzilla can be easily used in<br> + non-English speaking locales.<br> +<br> + To a degree field customisation supports this, because administrators<br> + could specify their own fields names anyway. However, there will<br> + always be some basic facilities not covered by this, and it is<br> + desirable that the administrator's interface also is<br> + internationalisable.<br> +<br> +Better Searching<br> +<br> + General Summary Reports<br> +<br> + Sometimes, the normal querying page leaves a lot to be desired. There<br> + are other facilities already in place or which people have asked for:<br> +<br> + Most Doomed Reports - All Bugs or All Bugs In A Product, Categorised<br> + On Assignee, Shows and Counts Number of Bugs For Each Assignee<br> + Most Voted For Bugs - All Bugs, Categorised On Product, Shows Top Ten<br> + Bugs Voters Most Want Fixed<br> + Number of Open Bugs For An Assignee - Bug List, Categorised On<br> + Developers, Counts Number of Bugs In Category<br> +<br> + The important thing to realise is that people want categorised reports<br> + on all sorts of things - a general summary report.<br> +<br> + In a categorised report, you choose the subset of bugs you wish to<br> + operate on (similar to how you would specify a query), and then<br> + categorise them on one or more fields.<br> +<br> + For each category you display the count of the number of things in<br> + that category. You can optionally display the bugs themselves, or<br> + leave them out, just showing the counts. And you can optionally limit<br> + the number of things (bugs or subcategories) that display in each<br> + category.<br> +<br> + Such a mechanism would let you do all of the above and more.<br> + Applications of this mechanism would only be recognised once it was<br> + implemented.<br> +<br> + Related Bugs<br> +<br> + It would be nice to have a field where you could enter other bugs<br> + related to the current bug. It would be handy for navigation and<br> + possibly even finding duplicates.<br> +<br> + Column Specification Support<br> +<br> + Currently bug lists use the columns that you last used. This doesn't<br> + work well for "prepackaged queries", where you followed a link. You<br> + can probably add a column by specifying a sort column, but this is<br> + difficult and suboptimal.<br> +<br> + Furthermore, I find that when I want to add a column to a bug list,<br> + it's usually a one off and I would prefer it to go away for the next<br> + query. Hence, it would be nice to specify the columns that appear on<br> + the bug list (and general summary report) pages. The default query<br> + mechanism should be able to let you specify your default columns.<br> +<br> + Advanced Querying Redesign<br> +<br> + ?<br> +<br> +Keywords<br> +<br> + People have a need to apply tags to bugs. In the beginning, people<br> + placed designators in the summary and status whiteboard. However,<br> + these fields were not designed for that, and so there were many flaws<br> + with this system:<br> + * They pollute the field with information that was never intended to<br> + be present.<br> + * Removing them with a bulk change is a difficult problem that has<br> + too many pitfalls to implement.<br> + * You can easily get the capitalisation wrong.<br> +<br> + Then dependencies were introduced (when?), and people realised that<br> + they could use them for "tracking bugs". Again, dependencies were not<br> + designed for that, and so there were more flaws, albeit different<br> + ones, including:<br> + * They aren't really bugs, so it's difficult to distinguish issues<br> + from bugs.<br> + * They can pollute bugs counts, and you must somehow exclude them<br> + from queries.<br> + * There is a whole lot of useless information on them. They have an<br> + assignee but there is nothing to fix, and that person can get<br> + whined at by Bugzilla. They have target milestones which must be<br> + manually maintained. And so on.<br> +<br> + Finally, keywords were introduced (when?) for this purpose to remove<br> + the need for these two systems. Unfortunately, the simple keywords<br> + implementation was itself lacking in certain features provided by the<br> + two previous systems, and has remained almost unchanged since its<br> + inception. Furthermore, it could not be forseen that in large<br> + installations, the sheer number of keywords could become unwieldly and<br> + could lead to a movement back to the other systems.<br> +<br> + The keywords system was the right idea, however, and it remains so.<br> + Fixing the keywords system is one of the most important Bugzilla<br> + issues.<br> +<br> + Bringing Keywords Up To Par<br> +<br> + For the most part, keywords are very good at what they do. It is easy<br> + to add and remove them (unlike summary/whiteboard designators), we can<br> + simply see what issues are present on a bug (unlike tracking bugs),<br> + and we do not confuse bugs with issues (unlike tracking bugs).<br> +<br> + However, there are still some "regressions" in the keyword system over<br> + previous systems:<br> + * Users wish to view the "dependency forest" of a keyword. While a<br> + dependency tree is of one bug, a dependency forest is of a bug<br> + list, and consists of a dependency tree for each member of the bug<br> + list. Users can work around this with tracking bugs by creating a<br> + tracking bug and viewing the dependency tree of that tracking bug.<br> + * Users wish to specify the keywords that initially apply to a bug,<br> + but instead they must edit the bug once it has already been<br> + submitted. They can work around this with summary designators,<br> + since they specify the summary at reporting time.<br> + * Users wish to store or share a bug list that contains a keywords<br> + column. Hence they wish to be able to specify what columns appear<br> + in the bug list URL, as mentioned earlier. They can work around<br> + this using summary designators, since almost all bug lists have a<br> + summary column.<br> + * Users wish to be able to view keywords on a bug list. However<br> + often they are only interested in a small number of keywords.<br> + Having a bug list with a keywords column means that all keywords<br> + will appear on a bug list. This can take a substantial amount of<br> + space where a bug has a lot of keywords, since the table columns<br> + in Bugzilla adjust to the largest cell in that column. Hence<br> + users wish to be able to specify which keywords should appear in<br> + the bug list. In a very real sense, each keyword is a field unto<br> + itself. Users can work around this by using summary designators,<br> + since they keywords will share the space in the summary column.<br> + * Users wish to know when bugs with a specific issue are resolved.<br> + Hence they wish to be able to receive notifications on all the<br> + bugs with a specific keyword. The introduction a generic watching<br> + facility (also for things like watching all bugs in a component)<br> + would achieve this. Users can work around this by using tracking<br> + bugs, as dependencies have an existing way of detecting fixes to<br> + bug a bug was blocked by.<br> +<br> + Dealing With The Keyword Overload<br> +<br> + At the time of writing, the mozilla.org installation has approximately<br> + 100 keywords, and many more would be in use if the keywords system<br> + didn't have the problems it does.<br> +<br> + Such a large number of keywords introduces logistical problems:<br> + * It must be easy for someone to learn what a keyword means. If a<br> + keyword is buried within a lot of other keywords, it can be<br> + difficult to find.<br> + * It must be easy to see what keywords are on a bug. If the number<br> + of keywords is large, then this can be difficult.<br> +<br> + These lead some people to feel that there are "too many keywords".<br> +<br> + These problems are not without solutions however. It is harder to<br> + find a list of designators or tracking bugs than it is a list of<br> + keywords.<br> +<br> + The essential problem is it needs to be easy to find the keywords<br> + we're interested in through the mass of keywords.<br> +<br> + Keyword Applicability<br> +<br> + As has been previously mentioned, it is desirable for fields to be<br> + able to adapt to the values of other fields. This is certainly true<br> + for keywords. Many keywords are simply not relevant because of the<br> + bugs product, component, etc.<br> +<br> + Hence, by introducing keyword applicability, and not displaying<br> + keywords that are not relevant to the current bug, or clearly<br> + separating them, we can make the keyword overload problem less<br> + significant.<br> +<br> + Currently when you click on "keywords" on a bug, you get a list of all<br> + bugs. It would be desirable to introduce a list of keywords tailored<br> + to a specific bug, that reports, in order:<br> + * the keywords currently on the bug<br> + * the keywords not currently on the bug, but applicable to the bug<br> + * optionally, the keywords not applicable to the bug<br> +<br> + This essentially orders the keywords into three groups, where each<br> + group is more important than the previous, and therefore appears<br> + closer to the top.<br> +<br> + Keyword Grouping & Ordering<br> +<br> + We could further enhance both the global and bug specific keyword list<br> + by grouping keywords. We should always have a "flat" view of<br> + keywords, but other ways of viewing the keywords would be useful too.<br> +<br> + If keyword applicability was implemented, we could group keywords<br> + based on their "applicability condition". Keywords that apply to all<br> + bugs could be separated from keywords that apply to a specific<br> + product, both on the global keyword list and the keyword list of a bug<br> + that is in that product.<br> +<br> + We could specify groups of our own. For example, many keywords are in<br> + a mutually exclusive group, essentially like radio buttons in a user<br> + interface. This creates a natural grouping, although other groupings<br> + occur (which depends on your keywords).<br> +<br> + It is possible that we could use collapsing/expanding operations on<br> + "twisties" to only should the groups we are interested in.<br> +<br> + And instead of grouping keywords, we could order them on some metric<br> + of usefulness, such as:<br> + * when the keyword was last added to a bug<br> + * how many bugs the keyword is on<br> + * how many open bugs the keyword is on<br> +<br> + Opting Out Of Keywords<br> +<br> + Not all people are going to care about all keywords. Therefore it<br> + makes sense that you may wish to specify which keywords you are<br> + interested in, either on the bug page, or on notifications.<br> +<br> + Other keywords will therefore not bother users who are not interested<br> + in them.<br> +<br> + Keyword Security<br> +<br> + Currently all keywords are available and editable to all people with<br> + edit bugs access. This situation is clearly suboptimal.<br> +<br> + Although relying on good behaviour for people to not do what they<br> + shouldn't works reasonably well on the mozilla.org, it is better to<br> + enforce that behaviour - it can be breached through malice, accident<br> + or ignorance.<br> +<br> + And in the situation where it is desirable for the presence or absence<br> + of a keyword not to be revealed, organisations either need to be<br> + content with the divulgence, or not use keywords at all.<br> +<br> + In the situation where they choose to divulge, introducing the ability<br> + to restrict who can see the keyword would also reduce keyword<br> + overload.<br> +<br> + Personal Keywords<br> +<br> + Keywords join together a set of bugs which would otherwise be<br> + unrelated in the bug system.<br> +<br> + We allow users to store their own queries. However we don't allow<br> + them to store their own keywords on a bug. This reduces the<br> + usefulness of personal queries, since you cannot join a set of<br> + unrelated bugs together in a way that you wish. Lists of bug numbers<br> + can work, by they can only be used for small lists, and it is<br> + impossible to share a list between multiple queries.<br> +<br> + Personal keywords are necessary to replace personal tracking bugs, as<br> + they would not pollute the keyword space. Indeed, on many<br> + installations this could remove some keywords out of the global<br> + keyword space.<br> +<br> + In a similar vein and with similar effects, group keywords could be<br> + introduced that are only available to members of a specific group.<br> +<br> + Keyword Restrictions<br> +<br> + Keywords are not islands unto themselves. Along with their potential<br> + to be involved in the inter-field relationships mentioned earlier,<br> + keywords can also be related to other keywords.<br> +<br> + Essentially, there are two possibilities:<br> + * a set of keywords are mutually exclusive<br> + * the presence of a keyword implies another keyword must be present<br> +<br> + Introduction of the ability to specify these restrictions would have<br> + benefits.<br> +<br> + If mutually exclusive keywords were present on a bug, their removal<br> + would fix up the database, as well as reducing the number of keywords<br> + on that bug.<br> +<br> + In the situation where a keyword implies another keyword, there are<br> + two possiblities as to how to handle the situation.<br> +<br> + The first is automatically add the keyword. This would fix up the<br> + database, but it would increase the number of keywords on a bug.<br> +<br> + The second is to automatically remove the keyword, and alter queries<br> + so they pick up the first keyword as well as the removed keyword.<br> + This would fix up the database and reduce the number of keywords on a<br> + bug, but it might confuse users who don't see the keyword.<br> + Alternatively, the implied keywords could be listed separately.<br> +<br> +Notifications<br> +<br> + Every time a bug gets changed notifications get sent out to people<br> + letting them know about what changes have been made. This is a<br> + significant feature, and all sorts of questions can be raised, but<br> + they mainly boil down to when they should be sent and what they should<br> + look like.<br> +<br> + Changes You're Interested In<br> +<br> + As of version 2.12 users can specify what sort of changes they are<br> + interested in receiving notifications for. However, this is still<br> + limited. As yet there is no facility to specify which keywords you<br> + care about, and whether you care about changes to fields such as the<br> + QA contact changes.<br> + Furthermore, often an unnecessary comment will go along with a change,<br> + either because it is required, or the commenter is ignorant of how the<br> + new system works. While explaining why you did something is useful,<br> + merely commenting on what you did is not because that information is<br> + already accessible view "Bug Activity".<br> +<br> + Because of this unnecessary comment, a lot of changes that would<br> + otherwise not generate notifications for certain people do so, because<br> + few people are willing to turn off comments. One way to deal with<br> + this problem is to allow people to specify that their comments are<br> + purely explanatory, and that anyone who is not interested in the<br> + change will not be interested in the comment.<br> +<br> + Furthermore, one possible rationale for unnecessary comments is that<br> + the bug activity does not display on the normal page and hence it is<br> + difficult to cross reference comments and actions. Hence, it would be<br> + beneficial to be able to do this.<br> +<br> + Bugs You're Watching<br> +<br> + Currently to receive a notification about a bug you need to have your<br> + name on it. This is suboptimal because you need to know about a bug<br> + before you can receive notifications on it. Often you are interested<br> + in any bug with a field set to a specific value. For example, you<br> + might be interested in all bugs with a specific product, component or<br> + keyword.<br> +<br> + If someone could automatically receive notifications about these bugs,<br> + it would make everyone's lives easier. Currently the default assignee<br> + and QA contact for a component will automatically receive<br> + notifications for<br> +<br> + Question: This moves half way to a BCC.<br> +<br> + Bulk Changes<br> +<br> + A very useful feature of Bugzilla is the ability to perform an action<br> + on multiple bugs at once. However, this means that similar<br> + notifications are currently generated for each bug modified.<br> +<br> + This can result in a torrent of notifications that can annoy.<br> +<br> + Furthermore, since the bugs are all changed close to each other in<br> + time, it is easy for someone to mass delete all the notifications<br> + generated by a bulk change and miss an unrelated notification in the<br> + middle.<br> +<br> + These factors can lead to a tendency for people to delay bulk changes,<br> + or avoid them entirely. This is suboptimal.<br> +<br> + It would be better if a bulk change generated only one notification<br> + mail. This would vastly reduce the annoyance factor, and prevent<br> + accidental deletion of notifications.<br> +<br> + One problem with this change is that some people separate out<br> + notifications using filtering. This means that they would no longer<br> + be match parts of a bulk change under different filtering rules.<br> +<br> + One possibility to resolve this is to allow people to specify groups<br> + of bugs. All bugs within a group would go into the same<br> + notification. The filters could then distinguish the different bug<br> + groups.<br> +<br> + In any case, it is likely there would need to be a transition period<br> + to allow people to alter their filters.<br> +<br> +Nominations<br> +<br> + ?<br> +<br> +Linking Bugzilla Installations<br> +<br> + The first example of linking Bugzilla installations together has is<br> + the introduction of bug moving in version 2.12. However, it would be<br> + useful to be able to link installations in more ways.<br> + * Dependencies and other relationships between bugs in other<br> + installations. This is difficult because dependencies are<br> + synchronised on both bugs, so the installation that changes<br> + dependencies would need to communicate the new state to the other<br> + installation. It would also mean that relationships and<br> + notifications that refer to other bugs would need to communicate<br> + with the other installation.<br> + * References to bugs in other installations. Currently if you type<br> + "bug XXX" or "bug #XXX" where XXX is a number, you get an<br> + automatic hyperlink to that bug. It would be useful if you could<br> + say "YYY bug #XXX" where YYY is the name of another installation.<br> +<br> +Retirement<br> +<br> + ?<br> +<br> +Whiny Reports<br> +<br> + ?<br> +<br> + Group Redesign<br> +<br> + ?<br> +<br> + Hard Wrapping Comments<br> +<br> + Currently Bugzilla "hard wraps" its comments to a specific line size,<br> + similar to E-Mail. This has various problems:<br> + * The way it currently works, wrapping is done in the browser at<br> + submission time using a non-standard HTML extension not supported<br> + by some (uncommon) browsers. These browsers generate comments<br> + that scroll off the right side of the screen.<br> + * Because comments are of fixed width, when you expand your browser<br> + window, the comments do not expand to fit available space.<br> +<br> + It would be much better to move to a world of soft wrapping, where the<br> + browser wraps the text at display time, similar to a world processor.<br> + And as in a word processor, soft wrapping does not preclude the<br> + insertion of newlines.<br> +<br> + Hard wrapping is too entrenched into text E-Mail to fix, but we can<br> + fix Bugzilla without causing any problems. The old content will still<br> + be wrapped too early, but at least new content will work.<br> + </P +> </P ></DIV ><DIV @@ -123,7 +712,7 @@ WIDTH="33%" ALIGN="right" VALIGN="top" ><A -HREF="faq.html" +HREF="variants.html" >Next</A ></TD ></TR @@ -142,7 +731,7 @@ VALIGN="top" WIDTH="33%" ALIGN="right" VALIGN="top" ->The Bugzilla FAQ</TD +>Bugzilla Variants and Competitors</TD ></TR ></TABLE ></DIV |