Notes on Searching the Digitized Uzbek National Bibliography

General notes on searching the documents:

1. There are two layers to each document. The top layer (what you see) is the image of the original scan. The bottom layer (which you don’t see) is the searchable characters. The top layer and the bottom layer don’t always match perfectly. If you see a word, but can’t find it when searching, this is an example of a mismatch between the layers.
2. Because there is sometimes a mismatch between the layers, these documents are best suited to keyword searches – i.e. to finding out most of what has been published on a particular topic. They are not well suited to use as a corpus (counting the exact number of time a word appears).
3. In some cases, the top and the bottom layer may not be aligned. If you copy and paste a word from the PDF, but you only get part of the word, or if you copy and paste a word, but get the word next to it, these are examples of how the text may not be perfectly aligned. This should not affect searching.
4. Russian words will be found with more accuracy than Uzbek words
5. Uzbek words will be found with more accuracy than Kyrgyz, Kazakh, and Karakalpak words
6. Uzbek Cyrillic words will be found with more accuracy than Uzbek and other Latin words ( in the 1928-1930, 1932, 1937-1940 documents only; for tips for searching these documents, see "Tips for Searching Latin Uzbek" ).
7. The following characters are not differentiated:
   ў у
   Й И
   Ё Е
   This means that a search for Ё will bring up results for both Ё and Е and so on.
8. Most hyphenated words that appear at line breaks will be searchable (i.e. "респу-блика" will be found by searching for "республика"). However, there are a few cases where hyphenated words have been misrecognized and will not be found.
9. Portions of some documents are from bad quality scans which may not be searchable. Slightly (very slightly) better quality scans are available if you request them from library staff. However, these are not searchable. In addition to these general notes, there are also specific things to be aware of concerning the different "fonts" in our database of PDFs. For information about the peculiarities of these fonts, see below.

Notes for the 1946-1961, 1978-1992, end of 1998-2000 documents and all of the Kitob Sovet Uzbekistana Kitobi documents:

In addition to the general issues outlined above, the biggest issue here is the confusion between "И" and "Н". We highly recommend searching with both characters. For example, if you want to find the word "ишлаш" search for both "ишлаш" and "ншлаш". This may results in additional results.

The following characters are also confused with one another; however, the problem here is not as bad as the confusion between И and Н. If feasible, we recommend searching with both characters for the following pairs as well:

Notes on Uzbek Latin text in the 1928-1930, 1932, 1937-1940 documents:

Bibliographies from these years use Latin rather than Cyrillic characters. The Latin alphabet used, however, differs significantly from the modern Uzbek Latin alphabet, as do spelling conventions. The following chart compares the Uzbek Latin alphabet of this period with modern Uzbek Latin:

*All occurrences of lowercase ʙ appear in the scanned documents as uppercase B.

**This character is recognized in the documents as I i, or not recognized at all. Try I i in your searches and/or try omitting the character completely.

NOTE: Some of these documents also contain Tajik, Kazak, and other Latin alphabets

Tips for Searching Uzbek Latin Portions of the Text:

Spelling in these portions of the documents is inconsistent and differs significantly from modern Uzbek spelling. You may search Uzbek Latin portion of the text using your keyboard and the additional special characters available through the onscreen keyboard at the top right hand of the page

If your search for a particular term returns no results, try alternate spellings, especially alternative vowels. To learn more about vowel usage in Latin Uzbek of this period, consult the chart below:

Note:

• The vowels Ə ə and Ɵ ɵ appear in some documents, but are not recognized. They may be recognized as A a and O o, respectively, but recognition is inconsistent.
• A character that resembles a Cyrillic soft sign (ь) appears occasionally to represent I i. This character is not recognized. If you fail to a find word with an I i sound, try omitting that character and searching again.
• Modern Uzbek J j are represented in the text as both Ç ç and Ƶ ƶ. The latter is used primarily in Russian words to represent Cyrillic Ж ж.
• The "ng" sound occurs in the texts alternately as N̡ ᶇ and Ng ng.
• The characters 1, I, and i are often improperly recognized and confused. Try alternate spellings.
• Words that in modern Latin Uzbek, are spelled with O o are often spelled with A a in the documents (ex. modern "Bola" is spelled "Bala"; modern "Ost" is spelled "Ast"; modern "Oltin" is spelled "Altin").

---