It's not quite that simple. 1. Having texts to search is one thing--and Accordance has a generous quantity of them already (LXX, pseud., Philo, Josephus, NT, Ap Fthrs, NT pseud., all grammatically tagged). There are ways to search all of them simultaneously, though I'll leave that to others to explain. 2. Having all other texts from the designated period in digital form is another thing. They are all available digitally in TLG, but I doubt that the TLG folks would offer them freely to Accordance. 3. Those texts, however, even if available, are not grammatically tagged and without that tagging, they have much more limited value. You're talking about *years* (probably decades) of work by a team of scholars to tag them accurately. (And no, public domain wiki-style tagging doesn't cut it. All that does is create an enormous pile of data that scholars would still have to sift, revise, and correct.) 4. And then you talk of translations! Add a bunch more time to digitize or create translations that are useful for the intended purpose. The Loeb translations are often extraordinarily "loose" (well beyond a functional equivalent in many cases). Though they give a fairly good sense of what the original text says, it's quite a challenge to figure out how to get from Greek to English or vice versa. (*Some* of the oldest Loeb vols. are now in public domain, but the newest ones are not--and it is an ongoing series.) 5. And even in an idealistic world where we already had all of those texts in digitally tagged format, simply making them available in Accordance (or any other such search engine) does not a lexicon make. All that's been provided is the data on which a lexicon might be created. A decent lexicon is an enormously complex project. For just a hint at what's involved, read John Lee's fascinating book that draws on his work to create a new Moulton & Milligan (a project that has, unfortunately, died after decades of work--which is a good illustration of the enormous work involved in preparing a lexicon): *A History of NT Lexicography* (SBG, 8; NY: Peter Lang, 2003), ISBN-13: 978-0820434803. And really anything such as a "lexicon-substitute" is not likely to be anything like a good lexicon; providing only raw data is rarely (if ever) of genuine value to most people. Sampling a subset of the data is misleading and mastering all of the Koine subset is more than all but a few will ever accomplish (I certainly have not and would never pretend otherwise). That's why *good* lexicons are essential. It's far more realistic to learn to use BDAG well and also to learn how to supplement that standard NT lexicon with more specialized tools such as LN, LSJ, MM, etc.
(I don't intend, BTW, to discourage anyone from working with and learning from the entire corpus of Koine texts, only to caution that doing so without a good lexicon or as a substitute for one is neither realistic nor well founded.)
Rodney J. Decker, ThD
Professor of NT & Greek
Baptist Bible Seminary