The following answer is adapted from a posting to 'comp.specification.larch' by Horning (July 19, 1995) in response to a question by Leavens (July 18, 1995). The trouble starts with your question: there are no "undefined terms" in LSL. LSL is a language of total functions. There are no partial functions associated with operators, although there may be underspecified operators (i.e., operators bound to functions with different values in different models).

On the other hand, please make sure that you've built the entire distribution with the same configuration. Not clear ? After running the "configure" command with its options, you should ideally run "make clean". Otherwise e.g. clients/Motif-1.2/mwm/Makefile may be configured such that it looks for stuff that isn't there in lib/Xm/* .

You can get the MIT releases of the LSL checker using the world-wide web, starting at the following URL. http://www.sds.lcs.mit.edu/spd/larch/lsl.html You can also get the MIT release by anonymous ftp from the following URL. ftp://ftp.sds.lcs.mit.edu/pub/Larch/LSL/ Iowa State, the Larch/C++ group has made available later beta releases of the LSL checker that fix problems with its profligate use of space. You can get the sources for Unix and Windows 95 executables from the following URL.

One gets warning messages about undefined and undeclared variables when compiling Java classes that have programming errors, as in the example below. There are many cases of poor runtime programming that could potentially be identified at compile time, but the number and subtlety of the cases gets increasingly difficult to address.

Wireshark can only be linked with version 4.2.2 or later of UCD SNMP. Your version of Wireshark was dynamically linked with such a version of UCD SNMP; however, you have an older version of UCD SNMP installed, which means that when Wireshark is run, it tries to link to the older version, and fails. You will have to replace that version of UCD SNMP with version 4.2.2 or a later version.

Accounting treatment of HUD Guaranteed and Direct Debt" are discussed in "GAAP Flyer #4". HUD guaranteed or direct debt gets written off to Contributed Capital (FDS Line Item No. 503).

The main difference between LSL and a Larch BISL is that in LSL one specifies mathematical theories of the operators that are used in the pre- and postcondition specifications of a Larch BISL. Thus LSL is used to specify mathematical models and auxiliary functions, and the a Larch BISL is used to specify program modules that are to be implemented in some particular programming language.

The Larch Shared Language (LSL) (see [Guttag-Horning93], Chapter 4, and [Guttag-Horning-Modet90]) is a language for specifying mathematical theories. LSL is a kind of equational algebraic specification language [Guttag75] [Guttag-Horning78] [Goguen-Thatcher-Wagner78] [Ehrig-Mahr85] [Futatsugi-etal85] [Mosses96] [Loeckx-Ehrich-Wolf96]. That is, specifications in LSL mainly consist of first-order equations between terms.

Besides this FAQ, the best place to look is probably your own computer system. You should have a manual page for the LSL checker, if it's installed on your system. Try the Unix command man lsl to see it. You should also look for a directory (such as '/usr/larch/LSL') where the installation of LSL is found. In that directory, you will find a subdirectory 'Doc', where there is some documentation on the checker. See section 2.

The sections of an LSL trait are determined by the LSL grammar [Guttag-Horning-Modet90]. (See section 2.3 Where can I get a checker for LSL?, for a more recent grammar, which is found in the file 'Doc/lsl3_1.y', for version 3.1, and 'Doc/lsl3_2.y', for version 3.2.

LSL trait denotes a theory, which is a collection of true formulas (of sort Bool). This theory contains "the trait's assertions, the conventional axioms of first-order logic, everything that follows from them, and nothing else" (see [Guttag-Horning93], p. 37). For a brief introduction to these ideas, see Chapter 2 of [Guttag-Horning93]; for general background on equational logic, see [Ehrig-Mahr85] or [Loeckx-Ehrich-Wolf96].

Technically, no; all functions specified in LSL are total (see section 2.14 What is the meaning of an LSL specification?). Thus every operator specified in LSL takes on some value for every combination of arguments. What you can do is to underspecify such an operator, by not specifying what its value is on all arguments. For example, the operator head in the handbook trait List (see [Guttag-Horning93], p. 173) is underspecified in this sense, because no value is specified for head(empty).

No, you don't have to specify everything completely in LSL. It's a good idea, in fact, to only specify the aspects that are important for what you are doing. For example, if you want to reason about graph data structures, it's best to not (at first, anyway) try to specify the theory of graphs in great detail. Put in what you want, and go with that. (This won't hurt anything, because whatever you need to prove you'll be forced to add eventually.

According to Guaspari (posting to the larch-interest mailing list, on March 8, 1995), "the commonest 'purely mathematical' mistakes" in trait definitions occur when one uses structural induction "over constructors that don't freely generate a sort". To understand this, consider his example, which includes the handbook trait Set (see [Guttag-Horning93], page 167).

The first tip helps you write down an algebraic specification of a sort that is intended to be used as an abstract data type (see [Guttag-Horning78], Section 3.4, and [Guttag-Horning93], Section 4.9). The idea is to divide the set of operators of your sort into generators and observers. A constructor returns your sort, while an observer takes your sort in at least one argument and returns some more primitive sort (such as Bool). The tip (quoted from [Guttag-Horning93], p.

The most commonly-used handbook of LSL traits is Guttag and Horning's handbook ([Guttag-Horning93], Appendix A). This can be obtained by anonymous ftp with the LSL checker (see section 2.3 Where can I get a checker for LSL?). A hypertext version is on-line in http://www.research.digital.com/SRC/larch/toc.html A general resource for all known handbooks that are publically available is found on the world-wide web, at the following URL. http://www.cs.iastate.edu/~leavens/Handbooks.

You can get a LaTeX style file, 'larch.sty', and a macro file defining a bunch of mathematical symbols, 'larchmath.tex', by anonymous ftp from the following URL. ftp://ftp.cs.iastate.edu/pub/larch/tex The documentation for 'larch.sty' says that it is to be used with LaTeX 2.09. However, it can be used with LaTeX2e. To do so, put the following lines at the start of your LaTeX input.

If the page on your Web server is tracked by HitLens, but there is no page name that corresponds to it in the HitLens Tracking system, you will see an 'Undefined page' in the HitLens reports. You inserted the HitLens script into a few local pages using the HitLens wizard and uploaded them to your host. After this, you removed the HitLens code from these pages and pressed 'Apply' or saved changes on exit.

The default timezone has not been specified on the Application Settings page, update that field and try again.

A - You need to edit your options and set 'List Top' to a bigger number. Set it to the number of top referrers you want to show on your site. Then recalculate your stats and everything should be fine.

I'm using a feature from YUI like menu, menubars, autocomplete... and it's getting put under my Grid (or other Ext element)

Currently, PAPI only supports thread level measurements with kernel or bound threads. Each thread must manipulate and read its own counters. When a thread is created, it inherits no PAPI information from the calling thread. Support for this kind of functionality is being implemented.

the infrastructure level, the Handle System allows individual namespaces to be registered and later discovered by any application over the Internet. Namespaces registered under the Handle System are unique, and queries will always be directed to the responsible handle server for resolution. An LDAP server typically runs in isolation, with service referrals from one namespace to another performed at the peer level when available.

Quantity of requests. If there are typically more requests per unit time than a given server can handle, the load on that server can be reduced by spreading it across more servers at that particular site. If a site becomes overloaded, one or more "secondary" sites may be used to replicate the primary site and thus to offload it. Quantity of data: If a single namespace becomes too large to be housed on a single server, it can be distributed across multiple servers that make up a site.

The following makefile shows one way to use the Unix command make to help check LSL traits. It relies on properties of the Bourne shell under Unix and standard Unix make, so it would have to be adjusted to work on other systems. (Also, if you cut and paste this makefile, be sure to change the leading blanks to tab characters on the rule lines.) SHELL = /bin/sh LSL = lsl LSLFLAGS = .SUFFIXES: .lsl .lsl-ckd .lp .lsl.lsl-ckd: $(LSL) $(LSLFLAGS) $< 2>&1 | tee $ .lsl.

LSL trait is used to describe a mathematical theory. This could be used by a mathematician to simply formalize a theory, but more commonly the theory specified is intended to be used as the mathematical vocabulary for some BISL. Another common use is as a way of specifying input to the Larch Prover (LP). When used as mathematical vocabulary for some behavioral interface specification, one can identify some other common uses. Quite often one wants to specify the abstract values of some data type.

In LSL 3.1, you can write a universal quantifier within an LSL term by using \A, and an existential quantifier using \E. As an example, consider the following trait (from Leavens's Math handbook).

Yes, there actually is a version of "spiderweb" specialized for use with LSL. If you are really a fan of such fancy systems, you can find it by using the literate programming library's URL. http://www.desy.de/user/projects/LitProg.html However, we have found that using the "noweb" system is much easier for most people, and nearly as good. You can get noweb from the literate programming library (see above), or directly from the following URL. http://www.eecs.harvard.edu/~nr/noweb/intro.

Yes, Penix's "lsl2html" tool converts an LSL input file to HTML, so it can be browsed over the net. It can be found at the following URL. http://www.ece.uc.edu/~kbse/lsl2html/ Unfortunately, Penix's tool has a few problems that have never been fixed. Instead, you might want to use Leavens's tool "lcpp2html", which is available from the following URL. http://www.cs.iastate.edu/~leavens/lcpp2html.html