| Related sites for http://www.cl.cam.ac.uk/~mgk25/publ-tips/ |
| Wikipedia__IBM Encylopedia article, including history and details of corporate culture. | | Coastside_Networking Specializes in selling and purchasing used and refurbished Cisco and Juniper Networks hardware. Located in Pleasant Hill, California. | | Ullanta_Performance_Robotics A theater troupe in which the actors are autonomous robots. The troupe presents traditional plays, performance art pieces, and dances; and, for recreation, participates in RoboCup robotic soccer comp | | STR_Software AventX Product Suite offers automated email and fax delivery of business documents generated by a wide range of applications. Includes case studies and datasheets. | | Elrond\'s_Gopher_Node Hosts a local search engine. Also offers gopher homepages. | | Hal_3000 Gopher server (pygopherd) serving in gopher, http and wap. Content updated hourly for weather and news as well as various other files. Site includes Jughead, Veronica and Jugtail search interfaces. | | Jugtail An open source search engine system for the Gopher protocol. Jugtail, formerly known as Jughead, searches one server at a time whereas Veronica doesn't have this limitation. | | JumpJet Hosts a large archive of up-to-date files and software on Gopher along with an interesting introduction to the Gopherspace. | | Kostecke Hosts GN, an open source multi-protocol server for Gopher and the World-Wide Web. A GopherMail system able to retrieve Gopher pages via e-mail is also available. | | PyGS Open source Gopher server written in Python. No longer being developed. | | Quasi-Indefatigable_Xenolith One of the most active Gopher servers on the Internet by a Gopher enthusiast. Many strange topics. | | Rachael`s_corner Hosts many software and hardware guides about Amiga and Gopher on Amiga. A Gopher server software for this system is also available. | | RFC_1436__The_Internet_Gopher_Protocol Internet standard for a distributed document search and retrieval protocol. | | Spencer_Hunter\'s_Homepage Example of a Gopher emulation in HTML, online since 1995. Under the "About this gopher and myself" directory is the author's own Gopher manifesto, "Why gopher is superior to the Web." | | Super_Dimension_Fortress Personal GOPHERSPACE is available to all users as well as hundreds of UNIX utilities, games and networking utilities. | | Vertrauen_BBS Gopher support demo for Synchronet, an open source BBS server. This Gopher service allows mainly a read-only access to some USENET newsgroups and other discussion boards. | | Whois_search Directory provided by the sipb.mit.edu Gopher server for querying whois servers throughout the Internet. Due to a port limitation, it doesn't work with Firefox, use Lynx instead. | | Wikipedia__Gopher_(protocol) Origins, decline, current status and technical details of the Gopher protocol. | | The_World Public gopher server. Features OBI, the Online Book Initiative. | | Gamedev_org Game development discussion forum. | | Simple_Game_Engine Information, screenshots and downloads for Simple Game Engine (freeware) enabling to games using a definition language. | | Xtreme_Games Free tools and source code and an offer to sign up new developers for joint ventures for a general pooling of resources. | | Robotics_at_Brown_University Research topics include environment mapping and path planning. | | The_Complete_Visual_Step-by-Step_Series Microsoft Office computer training manuals, for instructor-led and self-paced training. Unlimited printing rights and low cost site license. | | Topix__Search_Engines_News News on the search engine industry continually updated from thousands of sources around the Net. | | Multi-threaded_Debugging_Techniques Describes a number of general purpose debugging techniques for multi-threaded applications. (April 23, 2007) | | Trix_DrawingCenter View, markup, measure, print and automatically compare drawing files in DWG, DXF, DWF, and HPGL and PDF formats. | | Al_Zimmermann\'s_Programming_Contests A series of programming contests that offer cash prizes. Any programming language may be used. These contests are conducted at irregular intervals, generally once or twice a year. | | Megabitz Personal newsgroup service provider, offering unlimited or block accounts as well as a free trial. | | Power_Usenet Personal newsgroup accounts. Limited and unlimited download offers with different retention rates. | | Worldusenet Access to uncensored newsgroups with dedicated servers that are compatible with AOL, C15, MSN, cable and DSL. A web access is also provided. | | X-Privat Provides free and corporate anonymous NNTP and other network services. A completely free Usenet access is available for text-only newsgroups. | | Program_Base Listing of shareware and freeware software sorted by category. Accepts PAD submissions from authors. | | Books_by_Patrick_Henry_Winston_and_Friends Patrick H. Winston, (co-)author of "On To Java, Third Edition", "On To Smalltalk", "On To C++", "On To C", Artificial Intelligence, and Lisp presents his book. | | Guasi The GUASI library implements a thread based generic asyncronous execution engine, to be used to give otherwise syncronous calls an asyncronous behaviour. | | MultiMedia_SoftWare 32 bit DLLs for Windows: component for MS Windows Mixer controlling, component for any windows controlling. [Commercial] | | Amethyst_ShadowFX A sun and shadow modelling program for architects and town planners, working with data from DWG/DXF files. | | ARCAD Architectural CAD system for the Linux operating system. | | ArCon A 3D CAD software for use in civil engineering, architecture, and interior design. | | Argos_Systems,_Inc_ Software for construction and design management, with CAD tools for wood and steel frameworks and premanufactured housing. |
|
LaTeX and PDF: Effective scientific electronic publishing_uacct = "UA-78103-1";urchinTracker();Effective scientific electronic publishingMarkus G. Kuhn, Computer Laboratory,University of CambridgeThis is a brief list of recommendations for authors of scientificpapers who make their work available online. It focuses in particularon producing high-quality PDF files with LaTeX and covers some othertechnical and typographic pitfalls.ContentsBe consistent with how you write your nameUse the LaTeX styles suggested by the conference organizersMake sure your online version has page numbers and reference informationThe first printed page should be page number1Use PDF as the distribution format for your online versionUse Type1 vector fonts for generating PDF files with TeXSet the information fields of the PDF fileUse the paper format of the printed version in the PDF fileUse appropriate graphics formats for figuresUse good software engineering for the document sourcesUse filenames that are meaningful in a broader contextValidate your HTML filesTypographic conventionsBe consistent with how you write your nameChoose an exact spelling of your name at the start of yourscientific career and use that and only that on all your publications.Do not change any part of your name. If you have a middle initial inyour name, then either use it always (preferred) or use it never, butavoid switching between the two possibilities. Otherwise, you will getsorted in bibliographic databases (Science Citation Index, etc.) undervarious different places like J DOE and JA DOE, which makes it moredifficult to locate your work.Use the LaTeX styles suggested by the conference organizersIf the conference proceedings will be published by Springer as Lecture Notes inComputer Science, then use the latest LLNCS LaTeX2emacro package provided by Springer to format your camera-ready copy.Read the authors’instructions carefully.Make sure your online version has page numbers and reference informationCamera-ready submission formats required by publishers often lackpage numbers or an indication of where this paper was published,because the publishers want to add this information themselves. If youput this camera-ready version on the Web, then people will print itout and forget where they downloaded it. If they then can’t find thereference information on the paper, they will not be able to quoteyour paper properly.Therefore, your own online version should differ from the submittedcamera-ready copy in these two aspects. The page numbers should beswitched on and the precise bibliographic reference of your papershould be included. Preferably put the reference information at thebottom of the first page, in a way that does not change the pagebreaks compared to the submitted camera-ready copy.Update your online copy once you receive all the precisemetadata (page numbers, ISBN, publication date, etc.) of the finalpublished paper version.Users of Springer’s LLNCS style can use the package butterma.sty to add the bibliographicreference to the first page of the online version. To make this work,the file should start like \documentclass[runningheads]{llncs} \usepackage{butterma} \idline{J.~Doe and E.~Muster (Eds.): Perfect Publishing, LNCS 9999} \setcounter{page}{101} %\renewcommand{\year}{1999} % just if you don't want the current year ... \maketitle \thispagestyle{electronic}where of course the text after \idline has to be replacedwith your reference and the page number after\setcounter{page} has to be adjusted to your first pagenumber. The line \thispagestyle{electronic} which follows\maketitle suppresses the page number on the first page,which was activated by the [runningheads] in the firstline. If your title is too long to fit into the running heads, then youshould provide a shorter one using \titlerunning and\authorrunning. Special thanks to Antje Endemann from Springerfor this description. My variant prebutterma.sty, which only addsthe \idline value at the bottom, can be used to annotatepreprints that are not yet published in LNCS and therefore do not yetneed a Springer copyright notice and a page-number range.If you make a paper that you submitted to a publisher availableonline, then read the publisher’s copyright conditions carefully. Mostscientific publishers now allow you to have your paper on your Webpage, but some require you to add a special copyright notice.The first printed page should be page number 1International Standard ISO 7144 (“Documentation — Presentation oftheses and similar documents”):“The numbering of pages shall run consecutively, including blank pages,also if a thesis is published in several volumes, in arabic numerals,beginning on the recto of the first printed leaf. The title-leaves arecounted but not numbered.”If your document comes with a table of contents or index and theprinted version is usually bound separately (e.g., a thesis, technicalreport, manual, book), then it is very convenient if the page numbersprinted in the document match exactly the page numbers displayed by anelectronic document viewer, such as Adobe Reader or ghostview. This ismost easily achieved, if, starting from the title page (the front ofthe first page that comes out of the printer), all pages are numberedconsecutively in Arabic numerals (1, 2, 3, ...). The LaTeX “article”and “report” styles do this. Avoid separate Roman numerals for frontmatter (the LaTeX “book” style uses these by default). Should thethesis presentation regulations of your institution disagree, you maywant to make those who wrote them aware of ISO 7144.Use PDF as the distribution format for your online versionAdobe’s Portable Document Format is today clearly the preferred format forpublishing formatted documents. PDF has several advantages over themore traditional Adobe PostScript format:PDF was specifically designed as a document distribution andarchive format, while PostScript is just a printer control language.PDF is more portable.PDF viewers and printing tools such as the AdobeReader, GhostScript,xpdf, and kpdf are freely available. They aremuch more widely deployed and user friendly than PostScript printersand viewers (especially on Microsoft platforms).The Shrink to fit function of Adobe Reader provides anelegant workaround for the non-ISOpaper size problem with which North Americans continue to plaguethe world.PDF has built-in per-page compression, therefore it isnot necessary to use extra compression and packaging tools suchas PKZIP.PDF allows authors to include URL hyperlinks (accessible from TeXvia HyperTeX and the hyperref LaTeX macros).PDF encodes photos very compactly using the JPEG algorithm,whereas they increase the size of PostScript Level 1 filestremendously.Some PDF Web-browser plugins support fast direct download ofindividual pages in documents.PDF files can be created, for example, with the “ps2pdf” toolincluded with ghostscript, or with Adobe’sAcrobatDistiller, by converting PostScript files into PDF.Ghostscript versions before6.5 lacked full Type1 font support for PDF. All non-standard Type1fonts were transformed into 720 dpi pixel fonts when writing a PDFfile. Make sure you use a recent version.Please do not package PDF files into ZIP files. They arealready compressed. Put them directly on your web server. ApplyingPKZIP in addition will not reduce the size significantly, but it willrender the convenient PDF plug-ins of web browsers useless. Make surethat your web server serves PDF files with the line“Content-Type: application/pdf” in the HTTP header.PdfTeX is aversion of TeX that can produce both DVI and PDF files as output. Itknows a number of additional commands to control the PDF output(adding URLs, embedding graphics, etc.). Pdftex does not allow toembed EPS files as this is possible with dvips, but EPS files can beconverted into PDF using ghostscript and the epstopdf script thatcomes with tetex. The usual way of including diagrams into TeXdocuments is to use xfig or for morecomplicated cases MetaPost inorder to generate an embedded PostScript file. Both these tools allowyou to include mathematical formulae into diagrams that will betypeset by LaTeX (in xfig, export to "Combined PS/LaTeX (both parts)"to get a pair of pstex/pstex_t files).I’ll discuss below some of the more important issues of generatingPDF with TeX.Some related information can be found in:PDF on WikipediaTimothy Van Zandt’s Notes onconverting TeX or LaTeX documents to PDFMichael Shell’s testflowis a test page for your PDF conversion procedure that contains all thefeatures that have in the past been reported to go potentially wrongwhen producing PDF from TeX (see the documentationfor a detailed list).UsingLaTeX to Create Quality PDF Documents over the World Wide WebHowto make a compact beautiful PostScript or PDF file from a TeX filePDFzone.COM has a list ofPDF tools (freeware and commercial)Use Type1 vector fonts for generating PDF files with TeXTeX (and LaTeX) traditionally used raster-graphic fonts produced byMetafont for a specific device resolution. Dvipsoriginally produced PostScript files containing 300 or 600 dpi rasterfonts, and so did the PDF files converted from that by ps2pdf orAcrobat Distiller. PDF viewers do usually a rather bad job whendisplaying device-dependent “Type3” raster fonts. Texts in rasterfonts are displayed slow on the screen and with no or suboptimalanti-alias filtering. Also, the “Type3” raster fonts inserted by dvipslack information about which character each glyph represents, whichinterferes badly with full-text search and copy&paste.You can check whether the output of dvips contains any Type3raster/bitmap fonts under Unix with the command grep '%DVIPSBitmapFont:' file.pswhich should produce no output if there are no bitmap fonts.Instead of Type3 (raster graphics) fonts, make sure any Postscriptfile that you produce for conversion into PDF uses onlyresolution-independent Type1 vector fonts.Fortunately, aconsortium of AMS, SIAM, IBM, Springer, Elsevier, BlueSky Research,and Y&Y Inc. arranged to make commercial high-qualityPostScript Type1 versions of both the ComputerModern fonts and AMSfonts for TeX freely available under the copyright of AMS.Dvips has used these resolution-independent TeX fonts by defaultfor a few years now. If you still use some pre-2005 version of dvips,you may have to use special command-line options such as dvips -Ppdf -G0 ...to get the desired Type1 fonts. Or better upgrade to a more recent TeXdistribution. [The -G0 was a workaround for an old bugin dvips that caused ligatures to disappear in some fonts, which alsogot fixed in more recent versions.]Make sure you configure the distiller to the “Subset fonts below100%” option. This will ensure that only fonts for which 100% of allcharacters are used in the document are included completely and thedistiller will remove font data for all unused characters from yourPDF file. This will keep your PDF files small.When you want to convert to PDF historic PostScript files that wereproduced with Computer Modern bitmap fonts, then try thepkfixtool to replace these fonts in the PostScript with their Type1equivalents.Set the information fields of the PDF fileIn PDF files, you can store the title, authors, and keywords of apaper in special information fields. This information can help searchengines to locate and present your paper more accurately.There are several ways to set this information:You can manually add it using the Adobe Acrobat product.You can add via dvips a specialPostScript command that will instruct the distiller or ps2pdf toolto set these fields correctly. The direct way to do this is by addingsomewhere near the beginning of a TeX or LaTeX document code like \special{! /pdfmark where {pop} {userdict /pdfmark /cleartomark load put} ifelse [ /Author (Markus G. Kuhn, Ross J. Anderson) /Title (Soft Tempest: Hidden Data Transmission Using Electromagnetic Emanations) /Keywords (compromising emanations, data security, eavesdropping) /DOCINFO pdfmark}to store the author, title, and keywords in the PDF fileautomatically.Under LaTeX, the hyperref macros (see the hyperref manual), allow you to do the same in a slightly moreconvenient form via package options, as in \usepackage[pdftitle={Soft Tempest: Hidden Data Transmission Using Electromagnetic Emanations}, pdfauthor={Markus G. Kuhn, Ross J. Anderson}, pdfkeywords={compromising emanations, data security, eavesdropping}]{hyperref}Warning: The LLNCS style currently (version 2.14,2004-08-14) still has a few bugs that change its behaviour (e.g., thefigure/table caption font size) slightly when hyperref is loaded.With pdftex, you can also use \pdfinfo{/Author (Markus G. Kuhn, Ross J. Anderson) /Title (Soft Tempest: Hidden Data Transmission Using Electromagnetic Emanations) /Keywords (compromising emanations, data security, eavesdropping)}Use the paper format of the printed version in the PDF filePDF files look best on the screen if the specified paper sizematches the one for which the layout was designed. Therefore, use theactual physical paper size of the published document in the PDF file.When a PDF file is printed, the page will always be centered, and ifthe Shrink oversizes pages to paper size or Expand smallpages to paper size function is used it is also guaranteed to fitthe output paper size.For users of the LNCS style: the paper size is 152 mm × 235 mmand correct alignment of the output relative to the upper left cornercan be achieved by instructing distiller to ps2pdf to use theCropBox parameters [92 112 523 778].British Standard BS 1413 defines a book page size of156 mm × 234 mm called “Metric royal octavo”. This is the bestclue I have found so far on where the LNCS format might have comefrom.There are several ways to achive this:You can directly insert a special command into the DVI file rightafter \begin{document}: \special{! /pdfmark where {pop} {userdict /pdfmark /cleartomark load put} ifelse [ /CropBox [92 112 523 778] /PAGES pdfmark} % LNCS page: 152x235 mmThis line causes dvips to insert a pdfmark instruction into the PostScript output, which will set inthe distiller the page size correctly.If you use the hyperref macros in LaTeX, you can achieve the same for elegantlyvia the package options \usepackage[pdfpagescrop={92 112 523 778},a4paper=false]{hyperref}Warning: The LLNCS style currently (version 2.14,2004-08-14) still has a few bugs that change its behaviour (e.g., thefigure/table caption font size) slightly when hyperref is loaded.In pdftex, you can also use the command \pdfpagesattr{/CropBox [92 112 523 778]} % LNCS page: 152x235 mmUse appropriate graphics formats for figuresChoose appropriate formats for included graphics. In particular:Whenever possible, use resolution-independent vector graphicformats (e.g., EPS, WMF) for diagrams and line drawings such asblock diagrams;flow charts;function plots;circuit diagrams;diagrams that include text labels.Use pixel-based raster-graphic formats (PNG, JPEG, TIFF, GIF) onlyfor figures in which the original source information has a fixedresolution, such asscreen shots;photos;scanned paper documents;computer-generated raster graphics.Before considering lossy compression formats, such as JPEG and GIF,for presenting scientific data, make sure you understand exactly whatinformation their encoders throw away.In particular:JPEG was only designed for storing photos that will be viewedunmodified by humans.Avoid (DCT or wavelett) transform-based photo-compressionalgorithms, such as JPEG, for storing non-photographic orcomputer-generated material (e.g., screen shots). These formats arenot suitable for storing images that contain only a small number ofdistinct colours, large areas of identical color and distinct edges.Avoid lossy compression formats such as JPEG if your imagerepresents authentic scientific data that readers may want to inspectclosely with image-processing functions (zooming in, contrastenhancement, etc.).Avoid GIF, unless you know that the image contains fewer than 256different colors.Warning: Normally, distiller and ps2pdf will apply theDCT-JPEG compression to any colour and grayscale raster image thatthey encounter in the input PostScript file. In many scientificpublications, especially those related to image processing andcompression, this JPEG compression can introduce unacceptableartifacts that distort the meaning of the image. You can avoid this byprocessing the output of pnmtops with my sed script nojpeg.sed, which adds a setdistillerparamscommand to the generated EPS file that deactivates JPEG compression inthe distiller for this image only.To use nojpeg.sed in the Makefiledescribed in the next section, simply use the replacement macro PNMTOPS=pnmtops -rle -noturn -nosetpage | sed -f nojpeg.sedUse good software engineering for the document sourcesEnsure that your document preparation becomes a traceable andrepeatable process, just as you should have learned to do withsoftware (think ISO 9000).Use a revision control system like Subversion (manual) or RCS in order to keep track of oldrevisions and prepublications. Make sure that you can alwaysregenerate the source of any version of your paper that you ever havegiven away.Document-preparation systems that use plain-text document formats,such as LaTeX, are particularly easy to use with version-controlsoftware.You may find that the diff and merge functions work best if you starta new line after each sentence.If you work on a paper with coauthors, then cooperate via therevision control system and not by sending around new revisions viaemail.This avoids confusions and makes sure no changes are accidentallylost.If you use other TeX macro packages such as llncs.cls in your document, always add the version number andrelease date as a comment to where you include the macros. Keep allrevisions of macro packages that you use in a revision controldatabase as well.To include the file name and the RCS revision of your currentdraft on every printout, use the rcs.sty orrcsinfo.sty package, which extracts the relevantinformation from the $Id: ...$ strings that version-control tools canupdate automatically.Write a Makefile. If you used other software to generate embeddedpostscript files that you use in your TeX file (for instancefig2dev, gnuplot, etc.), then add rules for how thissoftware is called to your Makefile. Your Makefile should be able toregenerate automatically all intermediate files from the earliestprocessing step involved.If you use photographic images, then store them in an appropriateefficient file format (JPEG for photos, PNG for computer-generatedimages and scientific data) and use Makefile rules to generate thetemporary huge embedded PostScript files automatically from these withthe netpbmtools. Do not make it necessary to archive the potentially very largePostScript files. If it is not convenient for you to erase allPostScript files from your harddisk every evening, then you are doingsomething very wrong!A typical Makefile for a paper with various embeddedimages will contain production rules such as: .SUFFIXES: .png .tif .gif .jpg .pbm .pgm .ppm .dvi .fig \ .eps .ps .tex $(SUFFIXES) PNMTOPS=pnmtops -noturn -nosetpage .png.eps: pngtopnm $< | $(PNMTOPS)> $@ .tif.eps: tifftopnm $< | $(PNMTOPS)> $@ .gif.eps: giftopnm $< | $(PNMTOPS)> $@ .jpg.eps: djpeg $< | pnmtops -noturn -nosetpage> $@ .pbm.eps .pgm.eps .ppm.eps: $(PNMTOPS) $<> $@ .dvi.ps: dvips -Ppdf -G0 $< .tex.dvi: rm -f $*.ps $*.pdf latex $< || { rm -f $*.dvi $*.aux $*.idx && false ; } while grep 'Rerun to get ' $*.log ; do \ latex $< || { rm -f $*.dvi $*.aux $*.idx && false ; } ; done -killall -USR1 xdvi xdvi.bin xdvi-xaw3d.bin 2>/dev/null || true .fig.eps: fig2dev -L eps $< $@ .fig.pstex: .fig.pstex_t: fig2dev -L pstex_t -p $*.pstex $< $*.pstex_t fig2dev -L pstex $< $*.pstex clean: rm -f *.dvi *.log *.bak *.aux *.bbl *.blg *.idx *.ps *.eps *~Archive the source plus PDF version of your paper in a long-termformat. Standards such as tar, gzip, ISO 9660, PDF, TeX, and CD-R aretoday well documented and so widely deployed that people will mostlikely still be able to read them without major problems in 100 yearsfrom now. Eventually write your archive onto a CD-R, preferably thesilver or gold looking ones using the phthalocyanine dye, which shouldkeep the data intact for many decades, if not centuries. Remember thatmagnetic media lasts hardly longer than 5–10 years. Ensure that yourinstitution has a long-term archive concept for the source and finalformatted version of all publications.Use filenames that are meaningful in a broader contextIt is a good idea to include an indication of where the paper ispublished (abbreviation for the conference or journal) and a mostsignificant title word in the filename. For instanceih98-tempest.tex is much more useful then justpaper.tex. Plan your filenames such that you and all yourlocal colleagues can have them nicely together in a single publicdirectory. No filename should be longer than 25 characters; preferablykeep them at less than 15 characters. Use only lowercase US-ASCIIletters, digits, hyphens, and a dot (only for the extension).Validate your HTML filesIf you publish a HTML version of your paper, then please check notonly whether it displays nicely with your current browser, but alsosend it through an SGML parser that grammatically validates your HTMLsyntax against the HTML 4.01document type definition. A validation service is available forinstance from W3C, or you caneasily install your own using nsgmls. Also perform a link check from time totime, as URLs are unfortunately not very stable.Typographic conventionsProfessional typesetting works slightly different from usingtypewriters or ASCII email. Make sure that you are well familiar withthese conventions. Lamport’s LaTeX User’s Guide provides a very briefintroduction is section 2.2.1. In particular, make sure you are awareofhow to use directional quotation marks;the differences in shape and use of hyphen (-), minus (−), en-dash(–) and em-dash (—);the fact that the default font used by TeX’s math mode wasdesigned only for use with single-letter variables and must not beused for writing words or multi-letter abbreviations;When using BibTeX, understand that it tries to change thecapitalization of titles to lowercase unless a word is protected bysurrounding {}. Therefore, protect all proper nouns (names) andabbreviations in this way in your BibTeX file.Here are some more typographic conventions that you may want toconsider:Capitalization of headlines. In the United States, it is acommon practice to capitalize the first letter of more words inheadlines and titles than in normal sentences. The style guides andauthor’s instructions of U.S. publishers, such as the IEEE, requirethis. On the other hand, in Britain, in many other English-speakingcountries, and in many international organizations, professionaltypographers use in headlines exactly the same capitalizationrules as in normal sentences, namely only the first word and propernouns or abbreviations are capitalized. I personally strongly preferthe British convention. It preserves more information (which word is aname) and causes far fewer problems with bibliographic databases (likeBibTeX), where the unnecessary U.S.-style capitalization has to beremoved for most bibliographic-reference styles. My advice is tofollow the requirements of publishers, but if there are none, do notperform any unnecessary capitalization in titles. In any case, alwaysremain consistent within a single document.Quantities and units. There are well-established rules fortypesetting units of measurements, which are described, for example,in NISTSP 811 and ISO31-0. In particular:numbers and unit symbols are separated by a no-break space;unit symbols are never written in italics, to distinguishthem from variables for physical quantities, which are in italics;indices of variables are only written in italics if they representanother variable, but not if they are just an abbreviation of a word;symbols for SI unitshave well-defined capitalization rules (prefix symbols are uppercasefrom mega upwards and lowercase from kilo downwards, unit symbolsstart with an uppercase letter only if the unit was named after aperson).Bad example: vmax=120Kph Good example: vmax = 120 km/hSpecial thanks to RobinFairbairns and LarsEngebretsen for useful suggestions.Further suggestions for this text are very welcome! Just mail me.created 1998-05-01 – last modified 2008-06-17 –http://www.cl.cam.ac.uk/~mgk25/publ-tips/ |
|
