FAQs

• Japanese (all encodings)
• Korean
• Chinese (simplified and traditional)
• Thai
• Danish
• Dutch
• French
• German
• Italian
• Spanish
• Norwegian
• Portuguese
• Swedish
• Czech
• Hungarian
• Polish
• Romanian
• Russian
• Arabic
• Greek
• Hebrew
• Turkish
• English

The right way. We know that Lotus Notes is a complex enterprise application that many companies use exclusively for email communication, scheduling and contact management. Lotus allows users to create custom databases for storing content of various types. Because of the complexity and variety of Lotus Notes data, the best way to process this information is with the native application.

Our software automates this interaction and allows us to extract more data and cleaner content than other solutions. Don’t let someone convert your Lotus Notes data to another format before processing.

We often work on more than a dozen cases at a time.  Our ability to be successful with multiple simultaneous projects comes down to how well we perform in three key areas.

• Flexible technology: Our architecture is scalable. This is one of the many benefits of virtualization. If new engagements stretch our existing capacity, we simply create more server clones on the fly.


• The best staff: Logik hires and trains the brightest multi-talented people. Our team consists of industry veterans, consultants and technologists with distinguished backgrounds. We recognize the importance of cross-training our employees, meaning Todd in Sales can help monitor projects, and Adam from R&D knows about our pricing structures.


• Smart management: Our project and account managers are used to tight deadlines and handling eDiscovery requests of all shapes and sizes. To manage this many simultaneous projects requires exceptional communication and organization skills.  We do both really well.

Attachments are extracted during indexing and treated independently, though their relation to parent emails are always maintained.  Similarly, embedded objects are extracted and treated as separate files while maintaining the connection to the parent document.

These are extracted recursively to any depth level.  We offer clients a variety of options for viewing these relationships.  The most common is to provide BegAttach and EndAttach fields. 

We also provide AttachRange, ParentID, and ChildIDs fields for many of our clients.  In addition to the standard fields used for identifying relationships, we provide a True/False field that signals when an embedded object is present.

Yes, our foreign language detection algorithms are able to capture document encoding and to identify multiple languages, even in the same document. 

We supply fielded information containing the document languages and a relative weighting metric based on detection certainty.  We’ll also provide a field populated with all languages detected across an entire family of documents.

We do not have any limitations that prevent searching other than file types that do not contain text that is extractable or can be OCR’d (pictures, some 3D modeling files).  We also advise caution when dealing with relational databases. 

Often these file types do not lend well to data processing and especially to tiffing.  Much of the content of these file types is determined by the context created by the structure and format of the database.  Additionally, there will always be certain file types that do not contain printable content.  We provide slip sheets for these documents that list the file name and if desired, the md5 hash value.

Our email threading technology enables your review teams to quickly and easily sort all of their email by conversation thread. You can sort by custodian, or within a custodian’s PST or just by conversation thread.

Sorting by the conversation thread alone is usually not a good idea if multiple PSTs are involved, so we provide the two other fields, EMAIL_SORT1 and 2. Using a combined field sort, you can sort conversations within folders within PSTs, and by custodian if that information is provided in the file path. Using email thread analysis is a good idea. When you find an interesting document, you can easily review the entire conversation in order and in context. Alternatively, use threading to quickly identify an entire conversation and the related group of documents as non-responsive if off topic.

Alternatively, use threading to quickly identify an entire conversation and the related group of documents as non-responsive if off topic.

Simlogik is a technology that can detect and group similar or near-duplicative files. Unlike de-duplication technologies that detect exact duplicates, Simlogik detects slight variations in documents.

Simlogik is Unicode friendly. Unlike other near-duplicate technologies, Simlogik is not limited to just ASCII encoded files.  This means Simlogik can detect and group similar Chinese-to-Chinese, Japanese-to-Japanese, and even Arabic-to-Arabic documents together.

Note that this does not mean Simlogik can group a Chinese document with a similarly worded Arabic document.

We extract and normalize document text from native files using GridLogik.  Then, our algorithm assigns a doc-by-doc match score based on overlapping regions of the extracted text.  Extracted text allows for similarities to be detected across file types (e.g., email body text that is pasted into a Word document).  Files which have match scores above a certain threshold are considered ‘connected’. 

Connected groups of documents are assigned an id number which allows Simlogik near-dupes to clump together when sorted in a document review tool like Concordance.  You can also use the id numbers to make real-time queries if using an SQL-based document review tool (e.g. “find all similar documents by 75% or more to THIS doc). We calculate a doc-level group number and a parent-level group number which is cascaded down to children.  The parent-level number allows children to remain grouped with near-dupe parents when the results are sorted.

• Emails: For email, we create an MD5hash value by stringing together the following fields: From + To + BCC + CC + Subject + Sent Date + Received Date + BodyText(minus header) + Attachment Names = [MD5hash]. If one character is off in this formula the MD5hash value will be different.  So, if an email is exactly the same as another email except one has BCC, each email will be kept, because the hash values will be different.


• Loose files: For loose files, eDocs, we create an MD5hash value by putting the entire file into the MD5hash algorithm and then out comes an MD5hash.  Only the content of the file is analyzed for the MD5hash, not the file name.  So, you could have two exact files with different names, but be exact duplicates of one another, because the file name doesn’t affect the MD5hash.

So let it be written, so let it be done.  Since Gridlogik tracks every single byte of every file it comes in contact with, re-populating duplicate documents is very easy to do.  We can also provide a detailed de-dupe audit log that details what was de-duped and why. 

This is very helpful for horizontal de-duplication, because the log will show that a different person’s email may not be included, because it was de-duped.