docs/unix-phil
view unix-phil.ms @ 46:98a1446744c7
now concrete page numbers in ref markers
| author | meillo@marmaro.de |
|---|---|
| date | Mon, 12 Apr 2010 10:52:59 +0200 |
| parents | ade392f024aa |
| children | b6ae4a8ab1d3 |
line source
1 .nr PS 11
2 .nr VS 13
3 .nr lu 0
4 .de CW
5 .nr PQ \\n(.f
6 .if t .ft CW
7 .ie ^\\$1^^ .if n .ul 999
8 .el .if n .ul 1
9 .if t .if !^\\$1^^ \&\\$1\f\\n(PQ\\$2
10 .if n .if \\n(.$=1 \&\\$1
11 .if n .if \\n(.$>1 \&\\$1\c
12 .if n .if \\n(.$>1 \&\\$2
13 ..
14 .ds [. \ [
15 .ds .] ]
17 .rn NH _N
18 .de NH
19 .if '\\$1'1' .sp 2v
20 .if '\\$1'1' .nr PS +2
21 ._N \\$1
22 .if '\\$1'1' .nr PS -2
23 ..
25 .am QP
26 .ps -1
27 ..
29 .\"----------------------------------------
31 .TL
32 .ps +4
33 Why the Unix Philosophy still matters
34 .AU
35 markus schnalke <meillo@marmaro.de>
36 .AB
37 .ti \n(.iu
38 This paper explains the importance of the Unix Philosophy for software design.
39 Today, few software designers are aware of these concepts,
40 and thus a lot of modern software is more limited than necessary
41 and makes less use of software leverage than possible.
42 Knowing and following the guidelines of the Unix Philosophy makes software more valuable.
43 .AE
45 .FS
46 .ps -1
47 This paper was prepared for the ``Software Analysis'' seminar at University Ulm.
48 Mentor was professor Schweiggert. 2010-04-05
49 .br
50 You may retrieve this document from
51 .CW \s-1http://marmaro.de/docs \ .
52 .FE
54 .NH 1
55 Introduction
56 .XS
57 \*(SN Introduction
58 .XE
59 .LP
60 The Unix Philosophy is the essence of how the Unix operating system,
61 especially its toolchest, was designed.
62 It is no limited set of fixed rules,
63 but a loose set of guidelines which tell how to write software that
64 suites well into Unix.
65 Actually, the Unix Philosophy describes what is common to typical Unix software.
66 The Wikipedia has an accurate definition:
67 .[
68 wikipedia
69 unix philosophy
70 .]
71 .QP
72 The \fIUnix philosophy\fP is a set of cultural norms and philosophical
73 approaches to developing software based on the experience of leading
74 developers of the Unix operating system.
75 .PP
76 As there is no single definition of the Unix Philosophy,
77 several people have stated their view on what it comprises.
78 Best known are:
79 .IP \(bu
80 Doug McIlroy's summary: ``Write programs that do one thing and do it well.''
81 .[
82 mahoney
83 oral history
84 .]
85 .IP \(bu
86 Mike Gancarz' book ``The UNIX Philosophy''.
87 .[
88 gancarz
89 unix philosophy
90 .]
91 .IP \(bu
92 Eric S. Raymond's book ``The Art of UNIX Programming''.
93 .[
94 raymond
95 art of unix programming
96 .]
97 .LP
98 These different views on the Unix Philosophy have much in common.
99 Especially, the main concepts are similar in all of them.
100 McIlroy's definition can surely be called the core of the Unix Philosophy,
101 but the fundamental idea behind it all, is ``small is beautiful''.
103 .PP
104 The Unix Philosophy explains how to design good software for Unix.
105 Many concepts described here, base on facilities of Unix.
106 Other operating systems may not offer such facilities,
107 hence it may not be possible to design software in the way of the
108 Unix Philosophy for them.
109 .PP
110 The Unix Philosophy has an idea of how the process of software development
111 should look like, but large parts of the philosophy are quite independent
112 from a concrete development process.
113 However, one will soon recognize that some development processes work well
114 with the ideas of the Unix Philosophy and support them, while others are
115 at cross-purposes.
116 Kent Beck's books about Extreme Programming are valuable supplemental
117 resources on this topic.
118 .PP
119 The question of how to actually write code and how the code should looks
120 like in detail, are out of focus here.
121 ``The Practice of Programming'' by Kernighan and Pike,
122 .[
123 kernighan pike
124 practice of programming
125 .]
126 is a good book that covers this topic.
127 Its point of view matches to the one of this paper.
129 .NH 1
130 Importance of software design in general
131 .XS
132 .sp .5v
133 \*(SN Importance of software design in general
134 .XE
135 .LP
136 Software design is the planning of how the internal structure
137 and external interfaces of a software should look like.
138 It has nothing to do with visual appearance.
139 If we take a program as a car, then its color is of no matter.
140 Its design would be the car's size, its shape, the locations of doors,
141 the passenger/space ratio, the available controls and instruments,
142 and so forth.
143 .PP
144 Why should software get designed at all?
145 It is general knowledge, that even a bad plan is better than no plan.
146 Not designing software means programming without plan.
147 This will pretty sure lead to horrible results.
148 Software that is horrible to use and horrible to maintain.
149 These two aspects are the visible ones.
150 Often invisible though, are the wasted possible gains.
151 Good software design can make these gains available.
152 .PP
153 A software's design deals with quality properties.
154 Good design leads to good quality, and quality is important.
155 Any car may be able to drive from A to B,
156 but it depends on the car's properties whether it is a good choice
157 for passenger transport or not.
158 It depends on its properties if it is a good choice
159 for a rough mountain area.
160 And it depends on its properties if the ride will be fun.
162 .PP
163 Requirements for a software are twofold:
164 functional and non-functional.
165 .IP \(bu
166 Functional requirements define directly the software's functions.
167 They are the reason why software gets written.
168 Someone has a problem and needs a tool to solve it.
169 Being able to solve the problem is the main functional goal.
170 It is the driving force behind all programming effort.
171 Functional requirements are easier to define and to verify.
172 .IP \(bu
173 Non-functional requirements are called \fIquality\fP requirements, too.
174 The quality of a software are the properties that are not directly related to
175 the software's basic functions.
176 Tools of bad quality often do solve the problems they were written for,
177 but introduce problems and difficulties for usage and development, later on.
178 Quality aspects are often overlooked at first sight,
179 and are often difficult to define clearly and to verify.
180 .PP
181 Quality is hardly interesting when the software gets built initially,
182 but it has a high impact on usability and maintenance of the software, later.
183 A short-sighted might see in developing a software, mainly building something up.
184 But experience shows, that building the software the first time is
185 only a small amount of the overall work.
186 Bug fixing, extending, rebuilding of parts \(en maintenance work \(en
187 does soon take over the major part of the time spent on a software.
188 And of course, the time spent actually using the software.
189 These processes are highly influenced by the software's quality.
190 Thus, quality must not be neglected.
191 However, the problem with quality is that you hardly ``stumble over''
192 bad quality during the first build,
193 although this is the time when you should care about good quality most.
194 .PP
195 Software design has little to do with the basic function of a software \(en
196 this requirement will get satisfied anyway.
197 Software design is more about quality aspects of the software.
198 Good design leads to good quality, bad design to bad quality.
199 The primary functions of the software will be affected modestly by bad quality,
200 but good quality can provide a lot of additional gain,
201 even at places where one never expected it.
202 .PP
203 The ISO/IEC\|9126-1 standard, part\|1,
204 .[
205 iso product quality
206 .]
207 defines the quality model as consisting out of:
208 .IP \(bu
209 .I Functionality
210 (suitability, accuracy, inter\%operability, security)
211 .IP \(bu
212 .I Reliability
213 (maturity, fault tolerance, recoverability)
214 .IP \(bu
215 .I Usability
216 (understandability, learnability, operability, attractiveness)
217 .IP \(bu
218 .I Efficiency
219 (time behavior, resource utilization)
220 .IP \(bu
221 .I Maintainability
222 (analyzability, changeability, stability, testability)
223 .IP \(bu
224 .I Portability
225 (adaptability, installability, co-existence, replaceability)
226 .LP
227 Good design can improve these properties of a software,
228 bad designed software likely suffers in these points.
229 .PP
230 One further goal of software design is consistency.
231 Consistency eases understanding, working on, and using things.
232 Consistent internal structure and consistent interfaces to the outside
233 can be provided by good design.
234 .PP
235 Software should be well designed because good design avoids many
236 problems during a software's lifetime.
237 And software should be well designed because good design can offer
238 much additional gain.
239 Indeed, much effort should be spent into good design to make software more valuable.
240 The Unix Philosophy shows a way of how to design software well.
241 It offers guidelines to achieve good quality and high gain for the effort spent.
244 .NH 1
245 The Unix Philosophy
246 .XS
247 .sp .5v
248 \*(SN The Unix Philosophy
249 .XE
250 .LP
251 The origins of the Unix Philosophy were already introduced.
252 This chapter explains the philosophy, oriented on Gancarz,
253 and shows concrete examples of its application.
255 .NH 2
256 Pipes
257 .XS
258 \*(SN Pipes
259 .XE
260 .LP
261 Following are some examples to demonstrate how applied Unix Philosophy feels like.
262 Knowledge of using the Unix shell is assumed.
263 .PP
264 Counting the number of files in the current directory:
265 .DS
266 .CW
267 .ps -1
268 ls | wc -l
269 .DE
270 The
271 .CW ls
272 command lists all files in the current directory, one per line,
273 and
274 .CW "wc -l
275 counts the number of lines.
276 .PP
277 Counting the number of files that do not contain ``foo'' in their name:
278 .DS
279 .CW
280 .ps -1
281 ls | grep -v foo | wc -l
282 .DE
283 Here, the list of files is filtered by
284 .CW grep
285 to remove all lines that contain ``foo''.
286 The rest equals the previous example.
287 .PP
288 Finding the five largest entries in the current directory.
289 .DS
290 .CW
291 .ps -1
292 du -s * | sort -nr | sed 5q
293 .DE
294 .CW "du -s *
295 returns the recursively summed sizes of all files in the current directory
296 \(en no matter if they are regular files or directories.
297 .CW "sort -nr
298 sorts the list numerically in reverse order (descending).
299 Finally,
300 .CW "sed 5q
301 quits after it has printed the fifth line.
302 .PP
303 The presented command lines are examples of what Unix people would use
304 to get the desired output.
305 There are also other ways to get the same output.
306 It's a user's decision which way to go.
307 .PP
308 The examples show that many tasks on a Unix system
309 are accomplished by combining several small programs.
310 The connection between the single programs is denoted by the pipe operator `|'.
311 .PP
312 Pipes, and their extensive and easy use, are one of the great
313 achievements of the Unix system.
314 Pipes between programs have been possible in earlier operating systems,
315 but it has never been a so central part of the concept.
316 When, in the early seventies, Doug McIlroy introduced pipes into the
317 Unix system,
318 ``it was this concept and notation for linking several programs together
319 that transformed Unix from a basic file-sharing system to an entirely new way of computing.''
320 .[
321 aughenbaugh
322 unix oral history
323 .]
324 .PP
325 Being able to specify pipelines in an easy way is,
326 however, not enough by itself.
327 It is only one half.
328 The other is the design of the programs that are used in the pipeline.
329 They need interfaces that allow them to be used in such a way.
331 .NH 2
332 Interface design
333 .XS
334 \*(SN Interface design
335 .XE
336 .LP
337 Unix is, first of all, simple \(en Everything is a file.
338 Files are sequences of bytes, without any special structure.
339 Programs should be filters, which read a stream of bytes from standard input (stdin)
340 and write a stream of bytes to standard output (stdout).
341 If the files \fIare\fP sequences of bytes,
342 and the programs \fIare\fP filters on byte streams,
343 then there is exactly one data interface.
344 Hence it is possible to combine programs in any desired way.
345 .PP
346 Even a handful of small programs yields a large set of combinations,
347 and thus a large set of different functions.
348 This is leverage!
349 If the programs are orthogonal to each other \(en the best case \(en
350 then the set of different functions is greatest.
351 .PP
352 Programs can also have a separate control interface,
353 besides their data interface.
354 The control interface is often called ``user interface'',
355 because it is usually designed to be used by humans.
356 The Unix Philosophy discourages to assume the user to be human.
357 Interactive use of software is slow use of software,
358 because the program waits for user input most of the time.
359 Interactive software requires the user to be in front of the computer.
360 Interactive software occupy the user's attention while they are running.
361 .PP
362 Now to come back to the idea of combining several small programs,
363 to have a more specific function.
364 If these single tools would all be interactive,
365 how would the user control them?
366 It is not only a problem to control several programs at once,
367 if they run at the same time,
368 it also very inefficient to have to control each of the single programs
369 that are intended to act as one large program.
370 Hence, the Unix Philosophy discourages programs to demand interactive use.
371 The behavior of programs should be defined at invocation.
372 This is done by specifying arguments to the program call
373 (command line switches).
374 Gancarz discusses this topic as ``avoid captive user interfaces''.
375 .[ [
376 gancarz unix philosophy
377 .], page 88 ff.]
378 .PP
379 Non-interactive use is, during development, also an advantage for testing.
380 Testing of interactive programs is much more complicated,
381 than testing of non-interactive programs.
383 .NH 2
384 The toolchest approach
385 .XS
386 \*(SN The toolchest approach
387 .XE
388 .LP
389 A toolchest is a set of tools.
390 Instead of having one big tool for all tasks, one has many small tools,
391 each for one task.
392 Difficult tasks are solved by combining several of the small, simple tools.
393 .PP
394 The Unix toolchest \fIis\fP a set of small, (mostly) non-interactive programs
395 that are filters on byte streams.
396 They are, to a large extend, unrelated in their function.
397 Hence, the Unix toolchest provides a large set of functions
398 that can be accessed by combining the programs in the desired way.
399 .PP
400 There are also advantages for developing small toolchest programs.
401 It is easier and less error-prone to write small programs.
402 It is also easier and less error-prone to write a large set of small programs,
403 than to write one large program with all the functionality included.
404 If the small programs are combinable, then they offer even a larger set
405 of functions than the single large program.
406 Hence, one gets two advantages out of writing small, combinable programs:
407 They are easier to write and they offer a greater set of functions through
408 combination.
409 .PP
410 But there are also two main drawbacks of the toolchest approach.
411 First, one simple, standardized interface has to be sufficient.
412 If one feels the need for more ``logic'' than a stream of bytes,
413 then a different approach might be of need.
414 But it is also possible, that he just can not imagine a design where
415 a stream of bytes is sufficient.
416 By becoming more familiar with the ``Unix style of thinking'',
417 developers will more often and easier find simple designs where
418 a stream of bytes is a sufficient interface.
419 .PP
420 The second drawback of a toolchest affects the users.
421 A toolchest is often more difficult to use.
422 It is necessary to become familiar with each of the tools,
423 to be able to use the right one in a given situation.
424 Additionally, one needs to combine the tools in a senseful way himself.
425 This is like a sharp knife \(en it is a powerful tool in the hand of a
426 master, but of no good value in the hand of an unskilled.
427 However, learning single, small tools of a toolchest is easier than
428 learning a complex tool.
429 And the user will already have a basic understanding of a yet unknown tool,
430 if the tools of a toolchest have a common, consistent style.
431 He will be able to transfer knowledge over from one tool to another.
432 .PP
433 Moreover, the second drawback can be removed to a large extend
434 by adding wrappers around the basic tools.
435 Novice users do not need to learn several tools, if a professional wraps
436 complete command lines into a higher-level script.
437 Note that the wrapper script still calls the small tools;
438 it is just like a skin around them.
439 No complexity is added this way.
440 But new programs can get created out of existing one with very low effort.
441 .PP
442 A wrapper script for finding the five largest entries in the current directory
443 could look like this:
444 .DS
445 .CW
446 .ps -1
447 #!/bin/sh
448 du -s * | sort -nr | sed 5q
449 .DE
450 The script itself is just a text file that calls the command line,
451 which a professional user would type in directly.
452 It is probably worth to make the program flexible on the number of
453 entries it prints:
454 .DS
455 .CW
456 .ps -1
457 #!/bin/sh
458 num=5
459 [ $# -eq 1 ] && num="$1"
460 du -sh * | sort -nr | sed "${num}q"
461 .DE
462 This script acts like the one before, when called without an argument.
463 But one can also specify a numerical argument to define the number of lines to print.
464 One can surely imagine even more flexible versions, however,
465 they will still relay on the external programs,
466 which do the actual work.
468 .NH 2
469 A powerful shell
470 .XS
471 \*(SN A powerful shell
472 .XE
473 .LP
474 The Unix shell provides the possibility to combine small programs into large ones.
475 But a powerful shell is a great feature in other ways, too.
476 For instance by being scriptable.
477 Control statements are build into the shell.
478 The functions, however, are the normal programs of the system.
479 Thus, as the programs are already known,
480 learning to program in the shell becomes easy.
481 Using normal programs as functions in the shell programming language
482 is only possible because they are small and combinable tools in a toolchest style.
483 .PP
484 The Unix shell encourages to write small scripts,
485 by combining existing programs, because it is so easy to do.
486 This is a great step towards automation.
487 It is wonderful if the effort to automate a task equals the effort
488 to do the task a second time by hand.
489 If this holds,
490 then the user will be happy to automate everything he does more than once.
491 .PP
492 Small programs that do one job well, standardized interfaces between them,
493 a mechanism to combine parts to larger parts, and an easy way to automate tasks,
494 this will inevitably produce software leverage.
495 Getting multiple times the benefit of an investment is a great offer.
496 .PP
497 The shell also encourages rapid prototyping.
498 Many well known programs started as quickly hacked shell scripts,
499 and turned into ``real'' programs, written in C, later.
500 Building a prototype first, is a way to avoid the biggest problems
501 in application development.
502 Fred Brooks explains in ``No Silver Bullet'':
503 .[
504 brooks
505 no silver bullet
506 .]
507 .QP
508 The hardest single part of building a software system is deciding precisely what to build.
509 No other part of the conceptual work is so difficult as establishing the detailed
510 technical requirements, [...].
511 No other part of the work so cripples the resulting system if done wrong.
512 No other part is more difficult to rectify later.
513 .PP
514 Writing a prototype is a great method for becoming familiar with the requirements
515 and to run into real problems early.
516 .PP
517 Prototyping is often seen as a first step in building a software.
518 This is, of course, good.
519 However, the Unix Philosophy has an \fIadditional\fP perspective on prototyping:
520 After having built the prototype, one might notice, that the prototype is already
521 \fIgood enough\fP.
522 Hence, no reimplementation, in a more sophisticated programming language,
523 might be of need, at least for the moment.
524 Maybe later, it might be necessary to rewrite the software, but not now.
525 By delaying further work, one keeps the flexibility to react on
526 changing requirements.
527 Software parts that are not written will not miss the requirements.
529 .NH 2
530 Worse is better
531 .XS
532 \*(SN Worse is better
533 .XE
534 .LP
535 The Unix Philosophy aims for the 90% solution;
536 others call it the ``Worse is better'' approach.
537 Practical experience shows, that:
538 .PP
539 (1) It is almost never possible to define the
540 requirements completely and correctly the first time.
541 Hence one should not try to; one will fail anyway.
542 .PP
543 (2) Requirements change during time.
544 Hence it is best to delay requirement-based design decisions as long as possible.
545 The software should be small and flexible as long as possible
546 to react on changing requirements.
547 Shell scripts, for example, are more easily adjusted as C programs.
548 .PP
549 (3) Maintenance work is hard work.
550 Hence, one should keep the amount of code as small as possible;
551 it should just fulfill the \fIcurrent\fP requirements.
552 Software parts that will be written in future,
553 do not need maintenance till then.
554 .PP
555 Starting with a prototype in a scripting language has several advantages:
556 .IP \(bu
557 As the initial effort is low, one will likely start right away.
558 .IP \(bu
559 As working parts are available soon, the real requirements can get identified soon.
560 .IP \(bu
561 When a software is usable and valuable, it gets used, and thus tested.
562 Hence problems will be found at early stages of the development.
563 .IP \(bu
564 The prototype might be enough for the moment,
565 thus further work on the software can get delayed to a time
566 when one knows better about the requirements and problems,
567 than now.
568 .IP \(bu
569 Implementing now only the parts that are actually needed at the moment,
570 introduces fewer programming and maintenance work.
571 .IP \(bu
572 If the global situation changes so that the software is not needed anymore,
573 then less effort was spent into the project, than it would have be
574 when a different approach had been used.
576 .NH 2
577 Upgrowth and survival of software
578 .XS
579 \*(SN Upgrowth and survival of software
580 .XE
581 .LP
582 So far it was talked about \fIwriting\fP or \fIbuilding\fP software.
583 Although these are just verbs, they do imply a specific view on the work process
584 they describe.
585 The better verb, however, is to \fIgrow\fP.
586 Creating software in the sense of the Unix Philosophy is an incremental process.
587 It starts with a first prototype, which evolves as requirements change.
588 A quickly hacked shell script might become a large, sophisticated,
589 compiled program this way.
590 Its lifetime begins with the initial prototype and ends when the software is not used anymore.
591 While being alive it will get extended, rearranged, rebuilt.
592 Growing software matches the view that ``software is never finished. It is only released.''
593 .[ [
594 gancarz
595 unix philosophy
596 .], page 26]
597 .PP
598 Software can be seen as being controlled by evolutionary processes.
599 Successful software is software that is used by many for a long time.
600 This implies that the software is needed, useful, and better than alternatives.
601 Darwin talks about: ``The survival of the fittest.''
602 .[
603 darwin
604 origin of species
605 .]
606 Transferred to software: The most successful software, is the fittest,
607 is the one that survives.
608 (This may be at the level of one creature, or at the level of one species.)
609 The fitness of software is affected mainly by four properties:
610 portability of code, portability of data, range of usability, and reusability of parts.
611 .\" .IP \(bu
612 .\" portability of code
613 .\" .IP \(bu
614 .\" portability of data
615 .\" .IP \(bu
616 .\" range of usability
617 .\" .IP \(bu
618 .\" reuseability of parts
619 .PP
620 (1)
621 .I "Portability of code
622 means, using high-level programming languages,
623 sticking to the standard,
624 and avoiding optimizations that introduce dependencies on specific hardware.
625 Hardware has a much lower lifetime than software.
626 By chaining software to a specific hardware,
627 the software's lifetime gets shortened to that of this hardware.
628 In contrast, software should be easy to port \(en
629 adaptation is the key to success.
630 .\" cf. practice of prog: ch08
631 .PP
632 (2)
633 .I "Portability of data
634 is best achieved by avoiding binary representations
635 to store data, because binary representations differ from machine to machine.
636 Textual representation is favored.
637 Historically, \s-1ASCII\s0 was the charset of choice.
638 For the future, \s-1UTF\s0-8 might be the better choice.
639 Important is that it is a plain text representation in a
640 very common charset encoding.
641 Apart from being able to transfer data between machines,
642 readable data has the great advantage, that humans are able to directly
643 read and edit it with text editors and other tools from the Unix toolchest.
644 .\" gancarz tenet 5
645 .PP
646 (3)
647 A large
648 .I "range of usability
649 ensures good adaptation, and thus good survival.
650 It is a special distinction if a software becomes used in fields of action,
651 the original authors did never imagine.
652 Software that solves problems in a general way will likely be used
653 for many kinds of similar problems.
654 Being too specific limits the range of usability.
655 Requirements change through time, thus use cases change or even vanish.
656 As a good example in this point,
657 Allman identifies flexibility to be one major reason for sendmail's success:
658 .[
659 allman
660 sendmail
661 .]
662 .QP
663 Second, I limited myself to the routing function [...].
664 This was a departure from the dominant thought of the time, [...].
665 .QP
666 Third, the sendmail configuration file was flexible enough to adapt
667 to a rapidly changing world [...].
668 .LP
669 Successful software adapts itself to the changing world.
670 .PP
671 (4)
672 .I "Reuse of parts
673 is even one step further.
674 A software may completely lose its field of action,
675 but parts of which the software is build may be general and independent enough
676 to survive this death.
677 If software is build by combining small independent programs,
678 then these parts are readily available for reuse.
679 Who cares if the large program is a failure,
680 but parts of it become successful instead?
682 .NH 2
683 Summary
684 .XS
685 \*(SN Summary
686 .XE
687 .LP
688 This chapter explained central ideas of the Unix Philosophy.
689 For each of the ideas, the advantages they introduce were explained.
690 The Unix Philosophy are guidelines that help to write more valuable software.
691 From the view point of a software developer or software designer,
692 the Unix Philosophy provides answers to many software design problem.
693 .PP
694 The various ideas of the Unix Philosophy are very interweaved
695 and can hardly be applied independently.
696 However, the probably most important messages are:
697 .I "``Keep it simple!''" ,
698 .I "``Do one thing well!''" ,
699 and
700 .I "``Use software leverage!''
704 .NH 1
705 Case study: \s-1MH\s0
706 .XS
707 .sp .5v
708 \*(SN Case study: \s-1MH\s0
709 .XE
710 .LP
711 The previous chapter introduced and explained the Unix Philosophy
712 from a general point of view.
713 The driving force were the guidelines; references to
714 existing software were given only sparsely.
715 In this and the next chapter, concrete software will be
716 the driving force in the discussion.
717 .PP
718 This first case study is about the mail user agents (\s-1MUA\s0)
719 \s-1MH\s0 (``mail handler'') and its descendent \fInmh\fP
720 (``new mail handler'').
721 \s-1MUA\s0s provide functions to read, compose, and organize mail,
722 but (ideally) not to transfer it.
723 In this document, the name \s-1MH\s0 will be used to include nmh.
724 A distinction will only be made if differences between
725 \s-1MH\s0 and nmh are described.
728 .NH 2
729 Historical background
730 .XS
731 \*(SN Historical background
732 .XE
733 .LP
734 Electronic mail was available in Unix very early.
735 The first \s-1MUA\s0 on Unix was \f(CWmail\fP,
736 which was already present in the First Edition.
737 .[ [
738 salus
739 quarter century of unix
740 .], page 41 f.]
741 It was a small program that either printed the user's mailbox file
742 or appended text to someone elses mailbox file,
743 depending on the command line arguments.
744 .[
745 manual mail(1)
746 .]
747 It was a program that did one job well.
748 This job was emailing, which was very simple then.
749 .PP
750 Later, emailing became more powerful, and thus more complex.
751 The simple \f(CWmail\fP, which knew nothing of subjects,
752 independent handling of single messages,
753 and long-time email storage, was not powerful enough anymore.
754 In 1978 at Berkeley, Kurt Shoens wrote \fIMail\fP (with capital `M')
755 to provide additional functions for emailing.
756 Mail was still one program, but now it was large and did
757 several jobs.
758 Its user interface is modeled after the one of \fIed\fP.
759 It is designed for humans, but is still scriptable.
760 \fImailx\fP is the adaptation of Berkeley Mail into System V.
761 .[
762 ritter
763 mailx history
764 .]
765 Elm, pine, mutt, and a whole bunch of graphical \s-1MUA\s0s
766 followed Mail's direction.
767 They are large, monolithic programs which include all emailing functions.
768 .PP
769 A different way was taken by the people of \s-1RAND\s0 Corporation.
770 In the beginning, they also had used a monolithic mail system,
771 called \s-1MS\s0 (for ``mail system'').
772 But in 1977, Stockton Gaines and Norman Shapiro
773 came up with a proposal of a new email system concept \(en
774 one that honored the Unix Philosophy.
775 The concept was implemented by Bruce Borden in 1978 and 1979.
776 This was the birth of \s-1MH\s0 \(en the ``mail handler''.
777 .PP
778 Since then, \s-1RAND\s0, the University of California at Irvine and
779 at Berkeley, and several others have contributed to the software.
780 However, it's core concepts remained the same.
781 In the late 90s, when development of \s-1MH\s0 slowed down,
782 Richard Coleman started with \fInmh\fP, the new mail handler.
783 His goal was to improve \s-1MH\s0 especially in regard of
784 the requirements of modern emailing.
785 Today, nmh is developed by various people on the Internet.
786 .[
787 ware
788 rand history
789 .]
790 .[
791 peek
792 mh
793 .]
795 .NH 2
796 Contrasts to monolithic mail systems
797 .XS
798 \*(SN Contrasts to monolithic mail systems
799 .XE
800 .LP
801 All \s-1MUA\s0s are monolithic, except \s-1MH\s0.
802 Although there might actually exist further, very little known,
803 toolchest \s-1MUA\s0s, this statement reflects the situation pretty well.
804 .PP
805 Monolithic \s-1MUA\s0s gather all their functions in one program.
806 In contrast, \s-1MH\s0 is a toolchest of many small tools \(en one for each job.
807 Following is a list of important programs of \s-1MH\s0's toolchest
808 and their function.
809 It gives a feeling of how the toolchest looks like.
810 .IP \(bu
811 .CW inc :
812 incorporate new mail (this is how mail enters the system)
813 .IP \(bu
814 .CW scan :
815 list messages in folder
816 .IP \(bu
817 .CW show :
818 show message
819 .IP \(bu
820 .CW next\fR/\fPprev :
821 show next/previous message
822 .IP \(bu
823 .CW folder :
824 change current folder
825 .IP \(bu
826 .CW refile :
827 refile message into different folder
828 .IP \(bu
829 .CW rmm :
830 remove message
831 .IP \(bu
832 .CW comp :
833 compose new message
834 .IP \(bu
835 .CW repl :
836 reply to message
837 .IP \(bu
838 .CW forw :
839 forward message
840 .IP \(bu
841 .CW send :
842 send prepared message (this is how mail leaves the system)
843 .LP
844 \s-1MH\s0 has no special user interface like monolithic \s-1MUA\s0s have.
845 The user does not leave the shell to run \s-1MH\s0,
846 instead he uses the various \s-1MH\s0 programs within the shell.
847 Using a monolithic program with a captive user interface
848 means ``entering'' the program, using it, and ``exiting'' the program.
849 Using toolchests like \s-1MH\s0 means running programs,
850 alone or in combination with others, also from other toolchests,
851 without leaving the shell.
853 .NH 2
854 Data storage
855 .XS
856 \*(SN Data storage
857 .XE
858 .LP
859 \s-1MH\s0's mail storage is a directory tree under the user's
860 \s-1MH\s0 directory (usually \f(CW$HOME/Mail\fP),
861 where mail folders are directories and mail messages are text files
862 within them.
863 Each mail folder contains a file \f(CW.mh_sequences\fP which lists
864 the public message sequences of that folder,
865 for instance the \fIunseen\fP sequence for new messages.
866 Mail messages are text files located in a mail folder.
867 The files contain the messages as they were received.
868 They are named by ascending numbers in each folder.
869 .PP
870 This mailbox format is called ``\s-1MH\s0'' after the \s-1MUA\s0.
871 Alternatives are \fImbox\fP and \fImaildir\fP.
872 In the mbox format all messages are stored within one file.
873 This was a good solution in the early days, when messages
874 were only a few lines of text and were deleted soon.
875 Today, when single messages often include several megabytes
876 of attachments, it is a bad solution.
877 Another disadvantage of the mbox format is that it is
878 more difficult to write tools that work on mail messages,
879 because it is always necessary to first find and extract
880 the relevant message in the mbox file.
881 With the \s-1MH\s0 mailbox format, each message is a separate file.
882 Also, the problem of concurrent access to one mailbox is
883 reduced to the problem of concurrent access to one message.
884 The maildir format is generally similar to the \s-1MH\s0 format,
885 but modified towards guaranteed reliability.
886 This involves some complexity, unfortunately.
887 .PP
888 Working with \s-1MH\s0's toolchest on mailboxes is much like
889 working with Unix' toolchest on directory trees:
890 \f(CWscan\fP is like \f(CWls\fP,
891 \f(CWshow\fP is like \f(CWcat\fP,
892 \f(CWfolder\fP is like \f(CWcd\fP and \f(CWpwd\fP,
893 \f(CWrefile\fP is like \f(CWmv\fP,
894 and \f(CWrmm\fP is like \f(CWrm\fP.
895 .PP
896 \s-1MH\s0 extends the context of processes in Unix by two more items,
897 for its tools:
898 .IP \(bu
899 The current mail folder, which is similar to the current working directory.
900 For mail folders, \f(CWfolder\fP provides the corresponding functionality
901 of \f(CWcd\fP and \f(CWpwd\fP for directories.
902 .IP \(bu
903 Sequences, which are named sets of messages in a mail folder.
904 The current message, relative to a mail folder, is a special sequence.
905 It enables commands like \f(CWnext\fP and \f(CWprev\fP.
906 .LP
907 In contrast to Unix' context, which is maintained by the kernel,
908 \s-1MH\s0's context must be maintained by the tools themselves.
909 Usually there is one context per user, which resides in his
910 \f(CWcontext\fP file in the \s-1MH\s0 directory,
911 but a user can have several contexts, too.
912 Public sequences are an exception, as they belong to a mail folder,
913 and reside in the \f(CW.mh_sequences\fP file there.
914 .[
915 man page mh-profile mh-sequence
916 .]
918 .NH 2
919 Discussion of the design
920 .XS
921 \*(SN Discussion of the design
922 .XE
923 .LP
924 This section discusses \s-1MH\s0 in regard to the tenets
925 of the Unix Philosophy that Gancarz identified.
927 .PP
928 .B "Small is beautiful
929 and
930 .B "do one thing well
931 are two design goals that are directly visible in \s-1MH\s0.
932 Gancarz actually presents \s-1MH\s0 in his book as example under the
933 headline ``Making \s-1UNIX\s0 Do One Thing Well'':
934 .[ [
935 gancarz
936 unix philosophy
937 .], page 125 ff.]
938 .QP
939 [\s-1MH\s0] consists of a series of programs which
940 when combined give the user an enormous ability
941 to manipulate electronic mail messages.
942 A complex application, it shows that not only is it
943 possible to build large applications from smaller
944 components, but also that such designs are actually preferable.
945 .LP
946 The various programs of \s-1MH\s0 were relatively easy to write,
947 because each of them is small, limited to one function,
948 and has clear boundaries.
949 For the same reasons, they are also good to maintain.
950 Further more, the system can easily get extended.
951 One only needs to put a new program into the toolchest.
952 This was done, for instance, when \s-1MIME\s0 support was added
953 (e.g. \f(CWmhbuild\fP).
954 Also, different programs can exist to do the basically same job
955 in different ways (e.g. in nmh: \f(CWshow\fP and \f(CWmhshow\fP).
956 .PP
957 If someone needs a mail system with some additionally
958 functions that are not available anywhere yet, he best expands a
959 toolchest system like \s-1MH\s0.
960 There he can add new functionality by simply adding additional
961 programs to the toolchest.
962 There he does not risk to break existing functionality by doing so.
964 .PP
965 .B "Store data in flat text files
966 is followed by \s-1MH\s0.
967 This is not surprising, because email messages are already plain text.
968 \s-1MH\s0 stores the messages as it receives them,
969 thus any other tool that works on \s-1RFC\s0\|2822 mail messages can operate
970 on the messages in an \s-1MH\s0 mailbox.
971 All other files \s-1MH\s0 uses are plain text, too.
972 It is therefore possible and encouraged to use the text processing
973 tools of Unix' toolchest to extend \s-1MH\s0's toolchest.
975 .PP
976 .B "Avoid captive user interfaces" .
977 \s-1MH\s0 is perfectly suited for non-interactive use.
978 It offers all functions directly and without captive user interfaces.
979 If, nonetheless, users want a graphical user interface,
980 they can have it with \fIxmh\fP or \fIexmh\fP.
981 These are graphical frontends for the \s-1MH\s0 toolchest.
982 This means, all email-related work is still done by \s-1MH\s0 tools,
983 but the frontend calls the appropriate commands when the user
984 clicks on buttons.
985 .PP
986 Providing easy-to-use user interfaces in form of frontends is a good
987 approach, because it does not limit the power of the backend itself.
988 The frontend will anyway only be able to make a subset of the
989 backend's power and flexibility available to the user.
990 But if it is a separate program,
991 then the missing parts can still be accessed at the backend directly.
992 If it is integrated, then this will hardly be possible.
993 An additional advantage is the possibility to have different frontends
994 to the same backend.
996 .PP
997 .B "Choose portability over efficiency
998 and
999 .B "use shell scripts to increase leverage and portability" .
1000 These two tenets are indirectly, but nicely, demonstrated by
1001 Bolsky and Korn in their book about the Korn Shell.
1002 .[
1003 bolsky korn
1004 korn shell
1005 .]
1006 Chapter\|18 of the book shows a basic implementation
1007 of a subset of \s-1MH\s0 in ksh scripts.
1008 Of course, this is just a demonstration, but a brilliant one.
1009 It shows how quickly one can implement such a prototype with shell scripts,
1010 and how readable they are.
1011 The implementation in the scripting language may not be very fast,
1012 but it can be fast enough though, and this is all that matters.
1013 By having the code in an interpreted language, like the shell,
1014 portability becomes a minor issue, if we assume the interpreter
1015 to be widespread.
1016 .PP
1017 This demonstration also shows how easy it is to create single programs
1018 of a toolchest software.
1019 Eight tools (two of them have multiple names) and 16 functions
1020 with supporting code are presented to the reader.
1021 The tools comprise less than 40 lines of ksh each,
1022 in total about 200 lines.
1023 The functions comprise less than 80 lines of ksh each,
1024 in total about 450 lines.
1025 Such small software is easy to write, easy to understand,
1026 and thus easy to maintain.
1027 A toolchest improves the possibility to only write some parts
1028 and though create a working result.
1029 Expanding the toolchest, even without global changes,
1030 will likely be possible.
1032 .PP
1033 .B "Use software leverage to your advantage
1034 and the lesser tenet
1035 .B "allow the user to tailor the environment
1036 are ideally followed in the design of \s-1MH\s0.
1037 Tailoring the environment is heavily encouraged by the ability to
1038 directly define default options to programs.
1039 It is even possible to define different default options
1040 depending on the name under which a program is called.
1041 Software leverage is heavily encouraged by the ease of
1042 creating shell scripts that run a specific command line,
1043 built of several \s-1MH\s0 programs.
1044 There is few software that so much wants users to tailor their
1045 environment and to leverage the use of the software, like \s-1MH\s0.
1046 .PP
1047 Just to make one example:
1048 One might prefer a different listing format for the \f(CWscan\fP
1049 program.
1050 It is possible to take one of the distributed format files
1051 or to write one yourself.
1052 To use the format as default for \f(CWscan\fP, a single line,
1053 reading
1054 .DS
1055 .CW
1056 scan: -form FORMATFILE
1057 .DE
1058 must be added to \f(CW.mh_profile\fP.
1059 If one wants this different format as an additional command,
1060 instead of changing the default, he needs to create a link to
1061 \f(CWscan\fP, for instance titled \f(CWscan2\fP.
1062 The line in \f(CW.mh_profile\fP would then start with \f(CWscan2\fP,
1063 as the option should only be in effect for a program that is called as
1064 \f(CWscan2\fP.
1066 .PP
1067 .B "Make every program a filter
1068 is hard to find in \s-1MH\s0.
1069 The reason therefore is that most of \s-1MH\s0's tools provide
1070 basic file system operations for mailboxes.
1071 It is the same reason because of which \f(CWls\fP, \f(CWcp\fP, \f(CWmv\fP,
1072 and \f(CWrm\fP aren't filters neither.
1073 \s-1MH\s0 does not provide many filters itself, but it is a basis
1074 to write filters for.
1075 An example would be a mail text highlighter,
1076 that means a program that makes use of a color terminal to display
1077 header lines, quotations, and signatures in distinct colors.
1078 The author's version of such a program is an awk script with 25 lines.
1080 .PP
1081 .B "Build a prototype as soon as possible
1082 was again well followed by \s-1MH\s0.
1083 This tenet, of course, focuses on early development, which is
1084 long time ago for \s-1MH\s0.
1085 But without following this guideline at the very beginning,
1086 Bruce Borden may have not convinced the management of \s-1RAND\s0
1087 to ever create \s-1MH\s0.
1088 In Bruce' own words:
1089 .[ [
1090 ware rand history
1091 .], page 132]
1092 .QP
1093 [...] but [Stockton Gaines and Norm Shapiro] were not able
1094 to convince anyone that such a system would be fast enough to be usable.
1095 I proposed a very short project to prove the basic concepts,
1096 and my management agreed.
1097 Looking back, I realize that I had been very lucky with my first design.
1098 Without nearly enough design work,
1099 I built a working environment and some header files
1100 with key structures and wrote the first few \s-1MH\s0 commands:
1101 inc, show/next/prev, and comp.
1102 [...]
1103 With these three, I was able to convince people that the structure was viable.
1104 This took about three weeks.
1106 .NH 2
1107 Problems
1108 .XS
1109 \*(SN Problems
1110 .XE
1111 .LP
1112 \s-1MH\s0 is not without problems.
1113 There are two main problems: one is technical, the other is about human behavior.
1114 .PP
1115 \s-1MH\s0 is old and email today is very different to email in the time
1116 when \s-1MH\s0 was designed.
1117 \s-1MH\s0 adapted to the changes pretty well, but it is limited, though.
1118 \s-1MIME\s0 support and support for different character encodings
1119 is available, but only on a moderate level.
1120 This comes from limited development resources.
1121 More active developers could quickly change this.
1122 But \s-1MH\s0 is also limited by design, which is the larger problem.
1123 \s-1IMAP\s0, for example, conflicts with \s-1MH\s0's design to a large extend.
1124 These design conflicts are not easily solvable.
1125 Possibly, they require a redesign.
1126 \s-1IMAP\s0 may be too different to the classic mail model,
1127 which \s-1MH\s0 covers, so that \s-1MH\s0 may never support it well.
1128 .PP
1129 The other kind of problem are human habits.
1130 In this world, where almost all \s-1MUA\s0s are monolithic,
1131 it is very difficult to convince people to use a toolbox style \s-1MUA\s0
1132 like \s-1MH\s0.
1133 The habits are so strong, that even people who understand the concept
1134 and advantages of \s-1MH\s0 do not like to switch,
1135 simply because \s-1MH\s0 is different.
1136 Unfortunately, the frontends to \s-1MH\s0, which could provide familiar look'n'feel,
1137 are quite outdated and thus not very appealing, compared to the modern interfaces
1138 of many monolithic \s-1MUA\s0s.
1140 .NH 2
1141 Summary \s-1MH\s0
1142 .XS
1143 \*(SN Summary \s-1MH\s0
1144 .XE
1145 .LP
1146 \s-1MH\s0 is an \s-1MUA\s0 that follows the Unix Philosophy in its design.
1147 It consists of a toolchest of small tools, each of them does one job well.
1148 The toolchest approach offers great flexibility to the user.
1149 It is possible to utilize the complete power of the Unix shell with \s-1MH\s0.
1150 This makes \s-1MH\s0 a very powerful mail system.
1151 Extending and customizing \s-1MH\s0 is easy and encouraged.
1152 .PP
1153 Apart from the user's perspective, \s-1MH\s0 is development-friendly.
1154 Its overall design follows clear rules.
1155 The single tools do only one job, thus they are easy to understand,
1156 easy to write, and good to maintain.
1157 They are all independent and do not interfere with the others.
1158 Automated testing of their function is a straight forward task.
1159 .PP
1160 It is sad, that \s-1MH\s0's differentness is its largest problem,
1161 as its differentness is also its largest advantage.
1162 Unfortunately, for most people their habits are stronger
1163 than the attraction of the clear design and the power, \s-1MH\s0 offers.
1167 .NH 1
1168 Case study: uzbl
1169 .XS
1170 .sp .5v
1171 \*(SN Case study: uzbl
1172 .XE
1173 .LP
1174 The last chapter took a look on the \s-1MUA\s0 \s-1MH\s0,
1175 which is an old and established software.
1176 This chapter covers uzbl, a fresh new project.
1177 Uzbl is a web browser that adheres to the Unix Philosophy.
1178 Its name comes from the \fILolspeak\fP word for ``usable'';
1179 it is pronounced identical.
1181 .NH 2
1182 Historical background
1183 .XS
1184 \*(SN Historical background
1185 .XE
1186 .LP
1187 Uzbl was started by Dieter Plaetinck in April 2009.
1188 The idea was born in a thread in the Arch Linux Forums.
1189 .[
1190 arch linux forums
1191 browser
1192 .]
1193 After some discussion about failures of well known web browsers,
1194 Plaetinck (alias Dieter@be) came up with a very sketchy proposal
1195 of how a better web browser could look like.
1196 To the question of another member, if Plaetinck would write that program,
1197 because it would sound fantastic, Plaetinck replied:
1198 ``Maybe, if I find the time ;-)''.
1199 .PP
1200 Fortunately, he found the time.
1201 One day later, the first prototype was out.
1202 One week later, uzbl had an own website.
1203 One month after the first code showed up,
1204 a mailing list was installed to coordinate and discuss further development.
1205 Then a wiki followed to store documentation and scripts that showed up on the
1206 mailing list and elsewhere.
1207 .PP
1208 In the, now, one year of uzbl's existence, it was heavily developed on various branches.
1209 Plaetinck's task became more and more to only merge the best code from the
1210 different branches into his main branch, and to apply patches.
1211 About once a month, Plaetinck released a new version.
1212 In September 2009, he presented several forks of uzbl.
1213 Uzbl, actually, opened the field for a whole family of web browsers with similar shape.
1214 .PP
1215 In July 2009, \fILinux Weekly News\fP published an interview with Plaetinck about uzbl.
1216 In September 2009, the uzbl web browser was on \fISlashdot\fP.
1218 .NH 2
1219 Contrasts to other web browsers
1220 .XS
1221 \*(SN Contrasts to other web browsers
1222 .XE
1223 .LP
1224 Like most \s-1MUA\s0s are monolithic, but \s-1MH\s0 is a toolchest,
1225 most web browsers are monolithic, but uzbl is a frontend to a toolchest.
1226 .PP
1227 Today, uzbl is divided into uzbl-core and uzbl-browser.
1228 Uzbl-core is, how its name already indicates, the core of uzbl.
1229 It handles commands and events to interface other programs,
1230 and also displays webpages by using \fIwebkit\fP as render engine.
1231 Uzbl-browser combines uzbl-core with a bunch of handler scripts, a status bar,
1232 an event manager, yanking, pasting, page searching, zooming, and more stuff,
1233 to form a ``complete'' web browser.
1234 In the following text, the term ``uzbl'' usually stands for uzbl-browser,
1235 so uzbl-core is included.
1236 .PP
1237 Unlike most other web browsers, uzbl is mainly the mediator between the
1238 various tools that cover single jobs.
1239 Therefore, uzbl listens for commands on a named pipe (fifo), a Unix socket,
1240 and on stdin, and it writes events to a Unix socket and to stdout.
1241 Loading a webpage in a running uzbl instance requires only:
1242 .DS
1243 .CW
1244 echo 'uri http://example.org' >/path/to/uzbl-fifo
1245 .DE
1246 The graphical rendering of the webpage is done by webkit,
1247 a web content engine.
1248 Uzbl-core is built around libwebkit.
1249 .PP
1250 Downloads, browsing history, bookmarks, and the like are not provided
1251 by the core itself, like they are in other web browsers.
1252 Uzbl-browser also only provides, so called, handler scripts that wrap
1253 external applications which provide the actual functionality.
1254 For instance, \fIwget\fP is used to download files and uzbl-browser
1255 includes a script that calls wget with appropriate options in
1256 a prepared environment.
1257 .PP
1258 Modern web browsers are proud to have addons, plugins, and modules, instead.
1259 This is their effort to achieve similar goals.
1260 But instead of using existing, external programs, modern web browsers
1261 include these functions.
1263 .NH 2
1264 Discussion of the design
1265 .XS
1266 \*(SN Discussion of the design
1267 .XE
1268 .LP
1269 This section discusses uzbl in regard of the Unix Philosophy,
1270 as identified by Gancarz.
1272 .PP
1273 .B "Make each program do one thing well" .
1274 Uzbl tries to be a web browser and nothing else.
1275 The common definition of a web browser is, of course, highly influenced by
1276 existing implementations of web browsers, although they are degenerated.
1277 Web browsers should be programs to browse the web, and nothing more.
1278 This is the one thing they should do.
1279 .PP
1280 Web browsers should not, for instance, manage downloads.
1281 This is the job download managers exist for.
1282 Download managers do primary care about being good in downloading files.
1283 Modern web browsers provide download management only as a secondary feature.
1284 How could they do this job better, than programs that exist only for
1285 this very job?
1286 And how could anyone want less than the best download manager available?
1287 .PP
1288 A web browser's job is to let the user browse the web.
1289 This means, navigating through websites by following links.
1290 Rendering the \s-1HTML\s0 sources is a different job, too.
1291 It is covered by the webkit render engine, in uzbl's case.
1292 Audio and video content and files like PostScript, \s-1PDF\s0, and the like,
1293 are also not the job of a web browser.
1294 They should be handled by external applications \(en
1295 ones which's job is to handle such data.
1296 Uzbl strives to do it this way.
1297 .PP
1298 Remember Doug McIlroy:
1299 .I
1300 ``Write programs that do one thing and do it well.
1301 Write programs to work together.''
1302 .R
1303 .PP
1304 The lesser tenet
1305 .B "allow the user to tailor the environment
1306 matches good here.
1307 There was the question, how anyone could want anything less than the
1308 best program for the job.
1309 But as personal preferences matter, it is probably more important to ask:
1310 How could anyone want something else than his preferred program for the job?
1311 .PP
1312 Usually users want one program for a specific job.
1313 Hence, whenever the task is, for instance, downloading,
1314 the same download manager should be used.
1315 More advanced users might want to have this download manager in this
1316 situation and that one in that situation.
1317 They should be able to configure it this way.
1318 With uzbl, one can use any download manager the user wants.
1319 To switch to a different one, only one line in a small handler script
1320 needs to be changed.
1321 Alternatively it would be possible to query the program to use by
1322 reading a global file or an environment variable, in the handler script.
1323 .PP
1324 Uzbl does neither have its own download manager nor depends on a
1325 specific one, hence uzbl's browsing abilities will not be lowered by having
1326 a bad download manager.
1327 Uzbl's download capabilities will be just as good as the ones of the best
1328 download manager available on the system.
1329 Of course, this applies to all of the other supplementary tools, too.
1331 .PP
1332 .B "Use software leverage to your advantage" .
1333 Uzbl is designed to be extended by external tools.
1334 These external tools are usually wrapped by small handler shell scripts.
1335 Shell scripts are the glue in this approach.
1336 They make the various parts fit together.
1337 .PP
1338 The history mechanism of uzbl shall be presented as an example.
1339 Uzbl is configured to spawn a script to append an entry to the history
1340 whenever the event of a fully loaded page occurs.
1341 The script to append the entry to the history is not much more than:
1342 .DS
1343 .CW
1344 #!/bin/sh
1345 file=/path/to/uzbl-history
1346 echo `date +'%Y-%m-%d %H:%M:%S'`" $6 $7" >> $file
1347 .DE
1348 \f(CW$6\fP and \f(CW$7\fP expand to the \s-1URL\s0 and the page title.
1349 .PP
1350 For loading an entry, a key is bound to spawn a load-from-history script.
1351 The script reverses the history to have newer entries first,
1352 then displays \fIdmenu\fP to let the user select an item,
1353 and afterwards writes the selected \s-1URL\s0 into uzbl's command input pipe.
1354 With error checking and corner case handling removed,
1355 the script looks like this:
1356 .DS
1357 .CW
1358 #!/bin/sh
1359 file=/path/to/uzbl-history
1360 goto=`tac $file | dmenu | cut -d' ' -f 3`
1361 echo "uri $goto" > $4
1362 .DE
1363 \f(CW$4\fP expands to the path of the command input pipe of the current
1364 uzbl instance.
1366 .PP
1367 .B "Avoid captive user interfaces" .
1368 One could say, that uzbl, to a large extend, actually \fIis\fP
1369 a captive user interface.
1370 But the difference to most other web browsers is, that uzbl is only
1371 the captive user interface frontend (and the core of the backend).
1372 Many parts of the backend are independent of uzbl.
1373 Some are distributed with uzbl, for some external programs,
1374 handler scripts are distributed,
1375 but arbitrary additional functionality can be added if desired.
1376 .PP
1377 The frontend is captive \(en that is true.
1378 This is okay for the task of browsing the web, as this task is only relevant
1379 for humans.
1380 Automated programs would \fIcrawl\fP the web.
1381 That means, they read the source directly.
1382 The source includes all the semantics.
1383 The graphical representation is just for humans to transfer the semantics
1384 more intuitively.
1386 .PP
1387 .B "Make every program a filter" .
1388 Graphical web browsers are almost dead ends in the chain of information flow.
1389 Thus it is difficult to see what graphical web browsers should filter.
1390 Graphical web browsers exist almost only to be interactively used by humans.
1391 The only case when one might want to automate the rendering function is
1392 to generate images of rendered webpages.
1394 .PP
1395 .B "Small is beautiful"
1396 is not easy to apply to a web browser, because modern web technology
1397 is very complex, hence the rendering task is very complex.
1398 Modern web browsers have to consist of many thousand lines of code,
1399 unfortunately.
1400 Using the toolchest approach and wrappers can split the browser into
1401 several small parts, tough.
1402 .PP
1403 As of March 2010, uzbl-core consists of about 3\,500 lines of C code.
1404 The distribution includes another 3\,500 lines of Shell and Python code,
1405 which are the handler scripts and plugins like a modal interface.
1406 Further more, uzbl uses functionality of external tools like
1407 \fIwget\fP and \fInetcat\fP.
1408 Up to this point, uzbl looks pretty neat and small.
1409 The ugly part of uzbl is the web content renderer, webkit.
1410 Webkit consists of roughly 400\,000 (!) lines of code.
1411 Unfortunately, small web render engines are not possible anymore
1412 because of the modern web.
1414 .PP
1415 .B "Build a prototype as soon as possible" .
1416 Plaetinck made his code public, right from the beginning.
1417 Discussion and development was, and still is, open to everyone interested.
1418 Development versions of uzbl can be obtained very simply from the code
1419 repository.
1420 Within the first year of uzbl's existence, a new version was released
1421 more often than once a month.
1422 Different forks and branches arose.
1423 They introduced new features, which were tested for suitability
1424 for the main branch.
1425 The experiences of using prototypes influenced further development.
1426 Actually, all development was community driven.
1427 Plaetinck says, three months after uzbl's birth:
1428 ``Right now I hardly code anything myself for Uzbl.
1429 I just merge in other people's code, ponder a lot, and lead the discussions.''
1430 .[
1431 lwn
1432 uzbl
1433 .]
1436 .NH 2
1437 Problems
1438 .XS
1439 \*(SN Problems
1440 .XE
1441 .LP
1442 Similar to \s-1MH\s0, uzbl, too suffers from being different.
1443 It is sad, but people use what they know.
1444 Fortunately, uzbl's user interface can look and feel very much the
1445 same as the one of the well known web browsers,
1446 hiding the internal differences.
1447 But uzbl has to provide this similar look and feel to be accepted
1448 as a ``normal'' browser by ``normal'' users.
1449 .PP
1450 Though, the more important problem is the modern web.
1451 The modern web is simply broken.
1452 It has state in a state-less protocol,
1453 it misuses technologies,
1454 and it is helplessly overloaded.
1455 The result are web content render engines that must consist
1456 of hundreds of thousands lines of code.
1457 They also must combine and integrate many different technologies,
1458 only to make our modern web accessible.
1459 Website to image converter are hardly possible to run without
1460 human interaction because of state in sessions, impossible
1461 deep-linking, and unautomatable technologies.
1462 .PP
1463 The web was misused to provide all kinds of imaginable wishes.
1464 Now web browsers, and eventually the users, suffer from it.
1467 .NH 2
1468 Summary uzbl
1469 .XS
1470 \*(SN Summary uzbl
1471 .XE
1472 .LP
1473 ``Uzbl is a browser that adheres to the Unix Philosophy'',
1474 that is how uzbl is seen by its authors.
1475 Indeed, uzbl follows the Unix Philosophy in many ways.
1476 It consists of independent parts that work together,
1477 while its core is mainly a mediator which glues the parts together.
1478 .PP
1479 Software leverage can excellently be seen in uzbl.
1480 External tools are used, independent tasks are separated
1481 in independent parts and glued together with small handler scripts.
1482 .PP
1483 As uzbl, more or less, consists of a set of tools and a bit
1484 of glue, anyone can put the parts together and expand it
1485 in any desired way.
1486 Uzbl is very flexible and customizable.
1487 These properties make it valuable for advanced users,
1488 but may keep novice users from using it.
1489 .PP
1490 But uzbl's main problem is the modern web, that makes it hard
1491 to design a sane web browser.
1492 Despite this bad situation, uzbl does a fairly good job.
1495 .NH 1
1496 Final thoughts
1497 .XS
1498 .sp .5v
1499 \*(SN Final thoughts
1500 .XE
1502 .NH 2
1503 Quick summary
1504 .XS
1505 \*(SN Quick summary
1506 .XE
1507 .LP
1508 good design
1509 .LP
1510 unix phil
1511 .LP
1512 case studies
1514 .NH 2
1515 Why people should choose
1516 .XS
1517 \*(SN Why people should choose
1518 .XE
1519 .LP
1520 Make the right choice!
1522 .nr PI .3i
1523 .rm ]<
1524 .de ]<
1525 .LP
1526 .de FP
1527 .IP \\\\$1.
1528 \\..
1529 .rm FS FE
1530 ..
1531 .ds CH "
1532 .bp
1533 .rs
1534 .sp .3i
1535 .TL
1536 References
1537 .LP
1538 .XS
1539 .sp .5v
1540 References
1541 .XE
1542 .sp 2v
1543 .nr PS -1
1544 .nr VS -1
1545 .[
1546 $LIST$
1547 .]
1548 .\".wh -1p
1549 .bp
1550 .PX