Questions and answers

Is anything uploaded?

No. Your PDFs are opened and read in your browser tab by PDF.js and, for scanned pages, Tesseract OCR, both served from this site. Nothing from your documents is sent anywhere. The site counts visits with Cloudflare Web Analytics, which receives the page address, the page you came from, your browser and page-load timing (Web Vitals), and works out your country from the connection; never your files or anything in them. See the privacy page.

What do I get for $15?

One case number, open for 90 days: every proof of claim you drop in, matched to your scheduled debts; the debts no one claimed; the court's date where it runs from the bar date; and a memo with your action on each claim, ready to print. It is sold to lawyers and the staff they supervise. See the sample case.

Does it work in my district?

It works for a Chapter 13 case in any district. For 13 courts it knows the court's claims rule or order; for D. Minn., S.D. Fla. (LF-76) and N.D. Ga. (the certification) the memo is laid out for the court. Elsewhere the memo is a plain action list.

What about scanned claims?

A page is listed as read when its text was searched and the check saw nothing else on it; that does not mean nothing else is there. These pages are named as not read: a page with fewer than 40 characters of text beyond the court's header stamp (in the forms it knows, however the page is turned, and a stamp OCR misread), such as a scanned page (a page on which nothing at all is drawn is usually read, as blank), a page whose text looks scrambled, and every page of an XFA form that is built from form data it does not read (an XFA form that also carries its own pages and form fields, and whose form data it can read, is read by those pages, and its form data is searched). A page whose text was read but that also shows a picture it sees (a scan, a photo, a signature, a logo, a stamp image, digits drawn as shapes, or text in a font whose characters it could not check against the font itself; a picture in colour or faint print counts as much as black on white, whatever text lies over it, visible or not) is listed as text read, picture not read, with the picture's share of the page. A font can also give the wrong characters for what the page shows without this check noticing. Marks smaller than about 0.04 square inch, small empty outlines drawn as lines, and thin lines such as rules and borders are not counted; a checkbox drawn as a font symbol, comb boxes, a seal, a barcode or a rounded box corner can be listed as a picture. Every page is looked at at no less than 72 dots an inch, a large page in pieces (a page over about 4,600 square inches is listed as not read in full). What is in the file but does not show on screen is looked at too: a comment or a layer that is hidden or shows only in print is drawn like the rest, and an image covered by a box drawn over it, made invisible, clipped away or outside the page's visible area is listed as not read, and so is an image covered by another image. A page with a font it could not load is listed as not read in full too. It offers to read those pages with OCR in the same tab. OCR reads the pictures (the page drawn without the text already read) and adds what it reads to the page's own text, never replacing it. A page OCR cannot read with confidence, even turned on its side or upside down, stays on the list, and so does a picture in which OCR finds too little text (unless it reads as a few confident words with no digit in them, such as a logo's name, every part of it covered), or a part of a picture in which it finds no confident words. A page OCR read is marked read by OCR: check the image yourself, since OCR can misread a number. The text a page has is always searched, whatever the page is listed as: the text layer, visible or not, its comments, form fields (and what a field shows, when that differs from the field's value), and the text of a page whose fonts scrambled it, for the digits that survive. So are text placed outside the page's visible area, the document's properties and its bookmarks, form field names and tooltips, alternate text, scripts and page labels, and an attached file that is plain text (UTF-8, or UTF-16 with a byte-order mark); another attached file, or an email with encoded parts, is listed as not read. An earlier version an incremental save keeps in the file is compared with the pages only by its page text and its number of pictures: the page text it holds that the pages no longer show is searched, and the version is listed as not compared in full, since its comments, form fields, pictures and properties are not compared. Digits set apart one by one with single spaces are read as one number, and a birth date in separate boxes next to a birth-date label is read as a date; digits split into boxes or cells set far apart can still be missed. It never says a document is clear: when nothing is flagged, it says nothing was flagged in the text it could read, lists the pages read and those not read, and asks you to look at every page yourself.

Does it file anything or give legal advice?

No. It points to what to check, with the rule each check rests on, and the attorney decides. It files nothing.

How do refunds work?

For 3 days after the payment, a Refund button at the end of the review refunds the $15. After that, if the review could not open or read the claims in your case, write to us within 30 days of the payment and we refund it.

Can my paralegal use the code on another computer?

Yes. The unlock code opens the case number it was bought for, in any browser, until its end date. Keep it with the case file.

What does the memo look like?

The sample memo, printed from the made-up sample case: D. Minn., S.D. Fla. (LF-76), N.D. Ga. and any other district (PDF). Or open the sample case in the review.

Something else? [email protected]