4.5 documentation problems on first readthrough

[LosD]

Problems/potential improvements found when reading through documentation for 4.5 (not all are really related to the change in version, though).
Lines are in source XML lines. Sections are in source XML ids.

• General: We should use captions for images much more (ideally no image should ever be without a caption).
• Chapter 1.1 line 56: "a data quality analysis" -> "Data quality analysis"
• Chapter 1.1 section what_is_data_profiling: It would maybe be good to mention the quick analysis gained from a supported version (the "What can you tell me about my data" wizard).
• Chapter 1.1 line 171: Line says "As of version 3". That may be a bit uninteresting now we're at 4.5?
• Chapter 1.2 line 23: Line says that a license file is needed. But that is only true for licensed editions. Maybe add that?
• Chapter 1.2 section adding_components: The Improve super category is mentioned as only containing Transformers, which is not true.
• Chapter 1.2 line 241: 'but also "Write" menu for analyzers that save output to a datastore'. This is a bit simplistic, many non-writers can save to datastore. Maybe change to something like '... for analyzers that only (primarily?) save output to a datastore'.
• Chapter 1.2 line 285: Wording is a little weird: "In the bottom part of the canvas, a help message is displayed instructing what needs to be done in current moment to build a valid job."
• Chapter 2.2 line 160: The training tool does not open a new dialog. It's result screen is the "dialog".
• Chapter 2.2 section DE_suppression: Rename not really completed, the text still talks about suppression, and the image is using the old name.
• Chapter 2.2 section UK_suppression: Rename not really completed, the text still talks about suppression, and the image is using the old name. Maybe mention that it can use output data streams?
• Chapter 2.2 section US_suppression: Rename not really completed, the text still talks about suppression. Maybe mention that it can use output data streams?
• Chapter 2.2 section table_lookup_transformer: Maybe add that a new datastore can easily be registered with the new ➕ button and update screenshot.
• Chapter 2.2 section national_identifiers: Screenshot outdated, still contains EAN (even highlighted, to really draw attention to itself :)).
• Chapter 2.3 section completeness_analyzer: Maybe mention that it can use output data streams?
• Chapter 2.4 section writers_create_staging: Maybe add that a new datastore can easily be registered and a new table can easily be generated with the new ➕ buttons and update screenshot.
• Chapter 5.1 line 115: We do not provide signed JARs for download, so we should probably not say that we do.
• Chapter 8.2: Screenshots should probably be upgraded to reflect that AnalyzerBeansConfiguration has been renamed to DataCleanerConfiguration.

[tomaszguzialek]

Whenever specifying a page number, it refers to the printed version of html_single.

• "What is a datastore?" seems to be a DataCleaner-specific section, but it is contained within "Background and concepts" section which is an introduction to data quality, so a more general chapter.
• "II. Analysis component reference" title can now be confused with the newly introduced component reference. Consider renaming.
• We have "National identifiers" components documented, but we are missing newly introduced "Product data" components.
• The screenshot of the dropzone is without HDFS support. We should update the screenshot and description.
• Page 9: We are refering to the menus at the top. They are not there anymore, they have been replaced with the component library on the left.
• Page 22: Description of the screenshot talks about the Gender information in the Equals filter while the screenshot is about countries.
• Page 27: Dedup screenshot shows group similarity measure as ".99". We have changed that to more readable 99%. Let's update the screenshot.
• Page 33: References to retired EasyDQ brand and links to EasyDQ.com website.
• Page 35: There is a section dedicated to Completeness Analyzer, but no mention of output data streams in there.
• Page 36: In the table explaining "Lower case patterns expand in size". It says 'abc' won't match 'abc'? I am confused.
• Page 38: Optional properties of Value Distribution are outdated. Top and bottom N values are deprecated and not exposed in the UI anymore. There is "Record drill-down" information added, though.
• Page 41: "Configuration file" abstract mentions that conf.xml is not modifiable. Actually, new datastores are saved directly in conf.xml.
• Page 42: "Datastores" chapter gives examples of datastores, but it actually covers almost all options. Let's make it complete then and include HDFS, HBase and Cassandra.
• Page 43: ElasticSearch index has no mention of Shield support.
• Page 60: Docs say that GUI cannot save variables to XML files. We have fixed that recently.
• Page 66: Code snippet uses a deprecated method getOutcome instead of getFilterOutcome.
• Page 68: There are examples of extracting specific results from job results. We should add an example with AnalyzerResultFuture (not only AnalysisResultFuture which is already there).