MaybeToday.org

I updated my WORDsearch versions comparison matrix last night to include the current release of the new WORDsearch 9. Hopefully, this will requiring further update very soon, as there are a couple of significant bugs in version 9.0.1.88 that I sure hope are cleared up quickly.

WORDsearch 9 was released Monday, roughly two years after the release of version 8. This is the third release of WORDsearch developed on the Bible Explorer platform for CROSS eBooks. Long-time WORDsearch users who have been waiting for a return of the search results management genius of the old Ref List will not find what they’re looking for, but some significant improvements have been made to version 8 nonetheless.

The biggest improvement, by far, was a complete reworking of the window syncing mechanism. In the two previous iterations of WS, as in Bible Explorer, each window (for books organized on a book-chapter-verse structure, such as Bibles, commentaries, outlines, etc) had a sync button, which could be turned on or off by the user – a general preference setting would determine which state the window originally opened in. Every book which was synced would cause all other synced books to follow along with it as it changed context.

In WS9, there are now four distinct sync groups to which any sync-capable window can be assigned, meaning that certain windows can be synced to each other, while other windows can be synced together in a different context. This multi sync-group functionality would be familiar to users of Logos. However, WORDsearch also introduced an optional driver/slave mode, similar to how Pradis managed its sync groups – but WS9 allows the user to manage the driver status directly from each window, instead of launching a dialog box. With a driver window defined in a sync group, only the driver window will cause the other windows synced to it to change context – the slave windows can be moved around in without affecting any other windows, and will then re-sync to the driver window when the driver window changes context.

Also new is the ability to designate a Bible window as an xref target window (again, familiar to Logos users). This is a major improvement over the unpredictability of WS8 when clicking links, and is made even more useful when the target window is joined to a sync group.

Another long-overdue improvement in WORDsearch is support for a NOT operator in search strings, which is entered as ANDNOT.

Searching, in general, has become a bit friendlier, because of several subtle changes to the Search dialog interface that make it easier to find and select books to search. The Search box finally supports the ability to type <> delimiters, which allows the user to use one or more Strong’s numbers in the main search string (as opposed to using the special "Strong’s #" search dialog, which will only accept a single number).

Also new are options to automatically include plurals of English nouns, and various forms of English verbs. A new Spelling Helper applet on the search dialog box can help eliminate bogus searches due to spelling errors, as well as generating lists of available words in the selected resources. These extra helps are not quite as robust as the similar features in QuickVerse, but they should prove to be useful.

WS9 introduces the concept of the Carousel, which allows the user to define a set of frequently-used books of particular types (Bibles, commentaries, and dictionaries, roughly speaking), which will always be available to "flip" to from a similar window type, with a single mouse-click or keystroke. As you flip from favorite to favorite using the Carousel, the contents of the books will attempt to sync to the preceding favorite, though it’s hard to do that consistently with dictionary-type books.

In prior versions of the WS Verse List, the entire text area was "hot" for triggering sync events, and any time you clicked on a verse or passage in a Verse List, every window with syncing turned on would move to the verse you’d selected in the VL. In WS9, only the reference itself is "hot" for syncing; you can select the text of a verse without triggering a window sync. Furthermore, VLs now have Sync buttons, meaning syncing can be turned off altogether, or a specific sync group can be chosen. Clicking a "hot" area (ref) in a VL that is un-synced will drive a "target Bible" window to display the verse in the context of the translation defined by the VL ref link, making it easy to find the range of the passage you may want to expand your verse result into.

The Cross Reference Explorer (XRE) has been markedly improved from it’s initial iteration in WS8. There is now a Cancel button to kill unintended searches. Multiple hits for a reference within a single page/section are now collapsed into a single entry in the search results, with a (#) indicating how many hits are available on the page. The tree control for results has also been improved to require less mouse-clicking to get to your results.

XRE search hits in books (not user docs, unfortunately) are now highlighted in the content pane (making the tool MUCH more useful), and the content pane now automatically scrolls so that the first hit is visible – usually.

Bible Notes are also now searchable by the XRE – functionality that was always selectable in WS8, but never worked. This is a great way to create your own inverse cross-reference resource.

One of the key capabilities lost in the transition from WS7 to WS8 two years ago was the ability to create links in documents to sections of books. This capability has returned in WS9, and it is extremely simple to do – unlike the multi-step process involving bookmarks in WS7. WS9 also provides a simple interface for creating links to Biblical passages within documents and Bible Notes using plain text for the link (as opposed to Biblical references, which continue to automatically generate links to Biblical passages).

Fans of Inductive Bible Study – among others – will appreciate the new ability to assign labels to any or all of the dozen highlighting colors available, though I believe highlighting would be far more useful for Bible study if it were supported within Verse Lists. Another nice touch in WS9 is that highlighting can be applied now as either traditional highlighting, or as colored underlining.

The Instant Verse Study tool can now be populated using a Collection, and the Copy button now makes it very clear that content is only being copied to the clipboard, requiring the user to paste it wherever he wants to use it. Collections are much easier to use in this version, and a Manage Collections window has been provided. A Library Manager tool allows the user to hide unwanted books, and the SmartLink scripture popups will no longer position themselves so low on screen they cut off text.

On the downside – and avoiding complaining about what the program doesn’t do – several significant problems either persist, or were introduced with this release.

Searching MyDocuments for text – one of the potential deal-makers that could set this program in a class by itself among its competitors in this respect – has gone from blowing up whenever it encounters a malformed HTML file, to blowing up whenever it encounters a PDF file that is not an HTML file. Needless to say, no PDF file is an HTML file, and if you’re like most people these days, and you have PDF files (which WS has supported since WS8 as library resources), then document searching is utterly broken for you. The original problem goes back at least to WS7 – probably to the earliest versions of Bible Explorer – and the function is rendered worse than useless at this point. Search will not even return results from CROSS books if there is a PDF file in the search path.

Similarly to the initial release of WS8, support for Personal Notes for non-Bible books has vanished from this release. When WS8 was released, even the icon for Notes was gone from the windows, but it is still there in WS9 – it just doesn’t do anything. It remains to be seen whether they will make a comeback in WS9, like they did in WS8, but there was a hue and cry two years ago when they disappeared, and I can’t imagine the user response will be any different this time around.

I’m finding changes to the Search Results display a bit hard to adapt to (I’ve been using WS9 in beta for several months). There’s more of a tree structure now, and there is a pronounced focus on how results are distributed among the books of the Bible. That’s all terrific in theory, but there is no way to segregate results from multiple translations – they are treated independently within the results for each book of the Bible, but I sorely miss being able to scroll through the full results of each queried translation independently. I also miss the hit numbering, which has disappeared from the left-hand column. I understand what they were trying to do, and I’d like to see them refine it, but this screen looks more like a rough draft than a finished piece of work.

Not exactly a big surprise coming out of Zondervan today, as they have announced plans to drop their Pradis Bible Study software. Not a big loss to the industry either, I dare say, as Pradis was pretty narrowly focused on Zondervan resources (most of which were exclusively available in Pradis format), and always struck me as more geared toward promoting the interests of Zondervan than that of the Christian community. That’s OK – no law against that – but don’t look for any tears to be shed in this poor corner of the world. Frankly, Pradis wasn’t a very good program, either, though it did have its virtues.

Zondervan has some decent titles that might now be made available to people on different platforms, though it will be interesting to see whether they choose instead to market through another exclusive channel.  Based on this press release, that company will apparently be Logos if they do indeed go that way – and a corresponding release from Logos makes clear that Zondervan will be the party pulling the strings in terms of packaging, pricing, etc.

Of more interest to me is some of the language used in the press release, and how it highlights some of the inherent problems in the eBook industry, and the Bible Study Software market in particular. Zondervan is clearly playing down the fact that its customers are losing the investment they’ve made in Pradis, and in the books they’ve purchased to use with the program.

In the title of the release, Zondervan says it is “Retiring Pradis Software Search Engine.” Search engine? This was a software platform, built around a proprietary book format, that certainly included a search engine as a core component, but can hardly be reduced to it.

The release goes on to say Zondervan is “moving away from the Pradis software it created and will license other search engines.” You could be excused for thinking they are going to let other companies write software that will work with their proprietary book format – potentially giving customers who have purchased licenses for those books a new and improved means to work with them – but it seems certain that what they are actually doing is obsoleting those licensed books, and working out a licensing deal with Logos (and perhaps others) to get them re-published in a new proprietary format.

The next paragraph really gets to the heart of the matter:

“We are going to make sure we, first and foremost, work with the many thousands of Pradis customers for a smooth transition to the new search engines,” said Zondervan’s Paul Engle, Senior Vice President and Publisher of Church, Academic and Reference Resources. ”Many of these people have been customers for a dozen years and we will make sure they are eligible for a discount to these new titles upon release.”

Of course, these are not “new titles” at all that we are primarily talking about (Zondervan will apparently be licensing some new titles to Logos, which clouds the issue). For the most part, these are titles that customers have paid for in good faith, and are now going to have to pay for again in order the use with the “new search engines.” Eligible for a discount, indeed! Especially since the Logos engine is free! Zondervan is apparently planning on re-selling (at a discount!) these books to customers who have already bought them, so that the customers can use them with a free reader!

Other than BibleWorks – who actually warn customers about this exact potentiality and discourage them from buying large electronic libraries, all the Bible Software publishers encourage customers to purchase large libraries of resources, claiming they are more cost-effective to buy that way. That’s a true claim – unless your Bible Software publisher goes out of business, and you’re stuck with an expensive library of books that are doomed to obsoletion as the overall computer industry marches forward in its incessant cycle of progress. When you buy a hardback book, you can put it on your shelf, and it can serve a couple generations, regardless of what happens to the publisher.

There are other benefits to buying works electronically – as I’ve attested to here on this site in the past – and I remain an advocate for Bible Study Software, but the industry needs a standard for book formatting that will allow customers to retain their investment in books as they move among software providers. A standard rich enough to provide publishers plenty of elbow room to differentiate their own eBook products from competitors publishing the same work is certainly possible, and there is more than ample space for software publishers to differentiate their offerings.

And perhaps even more important to the publishers than the legitimate rights of library investors to viable licenses is the 800 pound gorilla standing in the doorway. If this matter can’t be resolved satisfactorily between the Christian software industry and the intellectuals with a stake in the outcome, then it will be settled de facto between Google and Amazon, and the Bible Study Software industry will be worse off for that.

So, it’s back to school time. I remember July being a blur of Russian and English, but I have no idea what happened to August. I suspended my own school work during the month so I could circle back to some earlier material in my Principles of Biblical Studies course that I felt I had glossed over too lightly, and I managed to get outlines generated of the five books of the Pentateuch, but I didn’t get much further than that, and I’m now staring at a new month, and need to get back on-track with the lessons, so I can complete the course on time.

I was using some of the great resources available in WORDsearch to help me put those outlines together, and I got a little frustrated (not for the first time) with the limitations of the internal word processor (WYSIWYG HTML editor, really), in combination with the drawbacks of using an external word processor in conjunction with the Bible Study program, and I started getting itchy to find something that would work better than either of those two options.

So, I started hunting (again, not for the first time) for a stand-alone WYSIWYG HTML editor that would offer a little more functionality than the internal WORDsearch wp – primarily a ‘view source’ mode, though other GUI-based layout tools would also be useful. Yet again, I was not able to find anything that worked for me (though I’ve since begun looking seriously into amaryllissoftware’s Selida), so I instead started looking into building a web-based tool that I could run from my website, which I could then access and use from within WORDsearch by utilizing the internal IE-based browser! I spent quite a bit of time trying to figure out how to get one working effectively, and while it’s still far from ready-for-primetime, I was able to use my little webEdit tool to build my outline for Deuteronomy this week.

If I’d be willing to set my sights low, I could be done with the project quickly, but I keep coming up with additional functionality that I want to roll into it. I’m tempted to open up the app to the public (specifically, to the WORDsearch user community), but I would have to incorporate authentication so I could implement security. That’s the only way I could allow people to upload pictures to include in their docs. Another option would be to develop a version that users could host on their own computers running under Microsoft’s built-in http server service, but I’m using PHP to do a lot of the backend stuff, so folks would need to install PHP on their computers as well, and I haven’t had a lot of luck getting PHP to run properly under Windows with IIS – maybe it was just me. I need to get this behind me soon, though…

WORDSearch released a reverse-interlinear based on the HCSB last week, further extending their original language capabilities with this translation. I wish I liked the HCSB better as a translation, because I really like WORDSearch as a study environment, and this looks like a very useful tool.

As far as I know, this is only the 3rd interlinear commercially released in a “reversed” format, which orders the text according to the translation rather than the manuscript, making it easier for novices, in particular, to work with the interlinear text. It joins the two very popular reverse-interlinears released by Logos a couple years ago (the ESV, and the NT of the NRSV). The Logos tools, however, are only available as part of base packages (although they’re included in almost all of them now, which I don’t think was the case before – I could be wrong). The WORDSearch HCSB reverse-interlinear is available as a stand-alone purchase, and is very nicely “upgrade” priced for those who own the standard WORDSearch Greek and/or Hebrew interlinears. The list price does seem quite pricey for those who don’t qualify for an “upgrade,” however, though it’s pretty common to have some “sale” or another going on.

This is the first installment in a series I plan to write, performing a side-by-side assessment of WORDsearch, Logos, QuickVerse, and Pradis.

QuickVerse has two tools for searching Biblical text: an Analytical Greek search tool designed to work with a morphologically tagged Greek NT module, and a general search tool used for searching English language Bible, as well as all other books – including user-created books. The most recent versions of QV introduce a couple other specialized searching tools I’ll discuss below.

The Search Dialog:

The general search tool has three modes (selectable from a drop-down): Text, Phrase, and Verse Reference. Newer versions of QuickVerse have a fourth mode, Text in Titles, which was an optional parameter to Text mode searches in version 11, and which I never found any use for.

The dialog box includes a pop-up keyboard for “typing” with mouse clicks in various languages. Recent search terms can be recalled in the search box, but associated parameters (even including search mode) are not likewise recalled, limiting the usefulness of this time-saving feature.

There is no facility to search for Strong’s codes as such, but you can search Strong’s-tagged books with a text search for a code, such as for G2588.

Search Logic:

Text searches allow for combining words using the Boolean operators AND, OR, XOR, and NOT. An implicit OR is supplied if there is no operator between two terms (personally, I think programs should instead AND terms by default, so that using more search terms narrows results rather than expanding them, but that is just a preference). Operators must be entered in ALL CAPS, or by using logical symbols (&, |, X, !). There’s a pick button for those who want to stop typing and use the mouse to input the operators between words.

There is no NEAR operator, but the Search dialog includes a drop-down box where the user selects whether the Boolean logic is applied against terms within the same verse, the same chapter, or the same book. This is a poor substitute for a NEAR operator, and I can’t imagine wanting to search for two words, but only if they exist in the same book!

Terms can be grouped in parentheses for greater control, and both * and ? wildcards are accepted. Regular expressions are not supported.

Setting Scope:

Selecting the target books to search is done via an expandable tree control which shows the entire library – including User Books. Multiple books – including multiple Bibles – can be searched simultaneously.

Setting a filter for Bible ranges is done via a drop-down text box. The only default range options are All, Old Testament, and New Testament, but you can type your own search ranges into the box, and recently defined custom ranges are added to the selectable options in the drop-down. Creating a search range for non-contiguous books is no problem (e.g. manually enter {Luke, Acts} to search only within Luke and Acts while ignoring John).

Extra-biblical text within Bibles (e.g. marginal notes and comments) cannot be searched, although – as noted below – book introductions cannot be excluded.

One peculiar search scope feature is an optional check box to limit the search to “Jesus words only” – a red-letter search, so to speak. I know for a fact that some people find this an important feature, though it strikes me as lending itself too easily to problematic views of Biblical inspiration and/or Christology.

Phrase Search:

Phrase Search mode takes all the text entered in the search box, and attempts to find the exact phrase within the selected books. This is “phrase searching for dummies,” an unfortunate approach that seriously limits phrase search functionality. Allowing the use of enclosing double-quotes in a Text search would be a much more robust solution.

Verse Reference:

In Verse Reference mode, the tool is actually a cross-reference lookup tool. In theory (and according to the Help file), this tool will not “search” Bibles for the verse arguments you give it, but is rather designed to search other types of books for occasions where the verse is mentioned in the text. In practice, I find that the tool can be used to retrieve the requested verses from every Bible in my library except the NAB. Not sure whether to file this under bug or feature.

This tool will accept multiple verses and/or ranges as arguments. References must be entered in a very specific format, using a colon (:) between the chapter and verse numbers. Also, a space must be placed between the leading number and the book name for books like 1 Jn (1Jn will not be understood). It can be tedious to have to abide by this nomenclature, but I’m sure you adapt to it if you use QV all the time.

Nice Special Features:

Search terms can be chosen from a word list that is available from the Search dialog, showing all words in each book selected as a search target.

Text searches include an option to search for related word forms, meaning that a search for {love} would return results for love, loves, loved, lover, loving, lovely, beloved, unloved, etc.

Another very neat option is the thesaurus search, which will include in results words with like or related meaning, even if they are not related lexically.

Analytical Greek Search:

The Analytical Greek search tool works with an available UBS4 text tagged with the Friberg morphology. Like the general tool, it supports Boolean operators, Bible ranges, and has word lists and a pop-up keyboard. It also has a morphology chart where you can select the grammatical parameters you want to search for or filter on. There are options to search lemma, and to exactly match diacritical marks, as well as to take everything entered as a single phrase term. This tool is quite serviceable. Unfortunately, there is no corresponding tool (or text) with which to perform analytical searching of the Hebrew.

New Searching Widgets:

Version 12 (2008) introduced a nice annotation search widget that will search all your annotations. Version 13 (2009) introduces a “Subject Search” feature which I haven’t seen yet, but will presumably function much like Topics searches in other programs.

Accuracy:

In the absence of wildcards, search will find only exact word matches (i.e. a search for {eat} will not return any hits on meat or heat, etc.). I think this is the right approach.

Text mode does not recognize strings entered within double-quotes as phrases, but instead treats each of the words as distinct search terms (placing an implicit OR between them). Because only Text mode supports multiple search terms, it is therefore not possible to perform a search such as {“Christ Jesus” AND became}, although you can always break the phrase apart into single words joined with AND logic: {Christ AND Jesus AND became}, which will produce close to the same results.

Searches against Biblical texts will return hits found in book introductions and prefaces, with, unfortunately, no way to limit the search to only the actual Bible text. You’d think an outfit that provided an option to search “red words only” might provide an option to search “inspired words only!”

Searches on words chosen from a word list sometimes return no results, even when searching against the entire library – which is a bit odd.

Performance:

Search performance is good on normal Text mode searches against a small number of books – often returning more or less immediately. Introducing some of the advanced options slowed things down somewhat – as expected. A search for {beauty} against 10 Bibles took about 3 seconds, but took about 7 seconds when doing a thesaurus search, and about 6 seconds when doing a related word forms search (and about 10 seconds when combining both features).

Introducing wildcards or multiple Boolean operators also began to drag down QuickVerse a bit, and it struggled somewhat searching a large number of books. Searching 10 Bibles for {gentile} took about 3 seconds – as expected. But this should be compared to my benchmark, WORDsearch, which performed the identical search in under 1 second.

Against the same Bibles, QuickVerse took six seconds to search for {gentile AND (God OR Lord)}, whereas WORDsearch took about 1 second.

Searching those same Bibles for {war* AND Lord AND House} took QuickVerse about 15 seconds, while WORDsearch again took about 1 second.

Searching a collection of about 125 books for the expression {war AND Lord} took QuickVerse 30 seconds, whereas WORDsearch performed the search against a similar number of books in about 4 seconds.

Verse Reference searches can be excruciatingly slow. The program seems to take forever to scan the Treasury of Scripture Knowledge (TSK) in particular, but struggles with many commentaries. The verse ref search is the kind of function that a user would be inclined to execute against the entire library. But if you did that, you could probably take the dog for a walk while waiting for it to return.

I removed TSK and a couple other books that seemed particularly problematic, and then ran the Verse Reference search for {Matt 24:22} against a collection of 15 books. The search took 1 minute and 40 seconds. Again for comparison, the same search against the exact same set of books in WORDsearch took less than 1 second.

In spite of these criticisms of search performance, it’s important to keep in mind that the majority of users will make the majority of their searches against one or perhaps a few Bibles, using simple search parameters – and QuickVerse will handle that fine.

Strengths in this area:

Word lists; fuzzy search features (thesaurus and word forms); XOR and NOT Boolean operators.

Weaknesses in this area:

Lack of Hebrew analytical search capability; lack of phrase support in multi-term text expression searches; lack of ability to search only Biblical text; overall searching performance.

One of my main goals for this site when I launched it last year was to provide assessments and comparisons of some of the Bible Study programs on the market – assessments based on what the various programs allow users to do, and how well they support those things, rather than focusing on the books available in different libraries.

I haven’t gotten very far to date, having started and stopped several times, and with course work looming on the horizon for me, my schedule is not going to be getting any looser. Given that, it seems like it’s now or never to get it going. I’ve decided to take an approach of publishing my results incrementally on the blog as I go through the process. I’m hoping this approach will provide me a little extra incentive to get through the job.

I’ll begin by focusing on the four commercial programs I currently have installed: WORDsearch, Logos, QuickVerse, and Pradis. Later on, I may add some or all of the other commercial, library-oriented programs to the mix (Accordance, PC Study Bible, PocketBible for Windows), but only if I can get evaluation copies – as I’m satisfied with the programs I’ve already made investments in, and don’t see the value in spending more money just to provide a public review of the other products.

I also hope to be able to provide evaluations of the programs I call text-oriented (BibleWorks, Bibloi, GRAMCORD), but this is also unlikely unless I can manage to get eval copies. I’m probably more likely to invest in one of these as a licensed study platform after evaluation, but that’s neither here nor there at this juncture.

As for the cohort of free programs out there, I may provide some comparative analysis later on, but it is not a priority – seeing that people can easily download and evaluate each one themselves. It would be nice if someone did the work of comparing them, and publishing the results, I’m just not sure that person will be me.

At least for starters, my plan is to evaluate each of the four programs more or less side-by-side, following the outline of evaluation criteria I published here. I’ll publish posts for each application’s evaluation at each stage of the eval process, beginning with the Core program functions – specifically: searching.

In between disasters and duties, I’ve been spending a bit more time looking at the new 2nd edition HCSB this week. Perhaps the most significant change from the first version is the greatly increased tendency to transliterate the Tetragrammaton (Yahweh), instead of following the standard practice among English translations of rendering it as LORD in small caps. Among major English translations, only the ASV (“Jehovah”) and the JB/NJB (“Yahweh”) have used a transliteration more than occasionally.

Curiously, the HCSB does not transliterate consistently, as the other translations had done (for reference, the other translations used the transliteration close to 7,000 times; the HCSB used the transliteration 78 times in the first edition, and uses it 495 times in the 2nd edition, if my search results can be trusted). The HCSB instead takes a more interpretive approach, transliterating whenever the usage suggests an emphasis on the word specifically as God’s name.

I have some misgivings about this approach – and I must admit that I am not crazy either about the ASV/JB-type approach of consistently transliterating, which I wouldn’t be very comfortable reading to a congregation. However, I’d like to think about it a little more before opining on the matter.

Nonetheless, this movement further in the direction of transliteration of the Tetragrammaton is clearly a significant part of the thinking behind the HCSB update, so I’d like to put it out there for anyone who might be interested in investigating exactly what they’ve done. There are 451 verses in the HCSB 2nd Edition that contain the word “Yahweh,” and that is too many for me to post here, so I will narrow the comparison down to the five books of the Torah.

The 2nd edition has 131 occurrences in 114 verses, whereas the 1st edition had just 13 occurrences in 12 verses. The table below is output from WORDsearch search results in the 2nd edition, and I have highlighted in yellow those 12 verse numbers where the 1st edition also transliterated: