Topics

Generational Search vs. Browse Behaviors #KM-research


 

Hello everyone,

I'm building an enterprise search strategy and am trying to prove or disprove the assumption that younger generations prefer search vs. browse in a consumer and enterprise information search context. I've attempted to research this lightly through our internal resources and am mostly finding content related to marketers or very old research. Is anyone aware of recent research on this topic that you could point me to, or is anyone currently conducting research here?

Thank you!
Rosanna Stephens


Daan Boom
 

Hi Rosanna:

Just received the attached in my box. Have not read it yet but it perhaps some of the questions you have on this subject.

Daan


On Apr 18, 2020, at 7:24 AM, Rosanna Stephens <rosannastephens@...> wrote:

Hello everyone,

I'm building an enterprise search strategy and am trying to prove or disprove the assumption that younger generations prefer search vs. browse in a consumer and enterprise information search context. I've attempted to research this lightly through our internal resources and am mostly finding content related to marketers or very old research. Is anyone aware of recent research on this topic that you could point me to, or is anyone currently conducting research here?

Thank you!
Rosanna Stephens


Matt Moore <innotecture@...>
 

Hi Rosanna,

Here are my comments (apologies if I am telling you what you already know)
- The consumer internet and enterprise information environments are very different and what is observed in one does not necessarily apply in the other.
- In Australia, every organisation that I have worked in, enterprise search has been poorly implemented (altho to be fair, one place did it OK for a while). There are organisations that do this well but they are the exception.
-  My suspicion is that information foraging behaviours within organisations are less driven by age and more by the quality of the tools available. i.e. whether or not you use search is more dependent on whether the search actually works rather than whether you fit into any generational category.

The best resources on enterprise search are Martin White's 2015 book and the surveys from Findwise (which sadly they have discontinued).

A quick Google yields a fair bit of research on how "young people" use search vs browsing on the internet - e.g. https://www.nngroup.com/articles/young-adults-ux/

Regards,

Matt

On Saturday, 18 April 2020, 09:24:30 am AEST, Rosanna Stephens <rosannastephens@gmail.com> wrote:

Hello everyone,

I'm building an enterprise search strategy and am trying to prove or disprove the assumption that younger generations prefer search vs. browse in a consumer and enterprise information search context. I've attempted to research this lightly through our internal resources and am mostly finding content related to marketers or very old research. Is anyone aware of recent research on this topic that you could point me to, or is anyone currently conducting research here?

Thank you!
Rosanna Stephens


Susan Ostreicher
 

Hi Matt, 

Thanks for the book and research recommendations - these are very helpful! 

I wondered if you could expand your first point. Do you mean that users prefer to act differently in the consumer internet vs. the enterprise? Or that they need to change their behavior because the two environments are so different? (Or something else?) Do you know of any research backing this up? 

On Fri, Apr 17, 2020 at 8:09 PM Matt Moore via groups.io <innotecture=yahoo.com@groups.io> wrote:
Hi Rosanna,

Here are my comments (apologies if I am telling you what you already know)
- The consumer internet and enterprise information environments are very different and what is observed in one does not necessarily apply in the other.
- In Australia, every organisation that I have worked in, enterprise search has been poorly implemented (altho to be fair, one place did it OK for a while). There are organisations that do this well but they are the exception.
-  My suspicion is that information foraging behaviours within organisations are less driven by age and more by the quality of the tools available. i.e. whether or not you use search is more dependent on whether the search actually works rather than whether you fit into any generational category.

The best resources on enterprise search are Martin White's 2015 book and the surveys from Findwise (which sadly they have discontinued).

A quick Google yields a fair bit of research on how "young people" use search vs browsing on the internet - e.g. https://www.nngroup.com/articles/young-adults-ux/

Regards,

Matt

On Saturday, 18 April 2020, 09:24:30 am AEST, Rosanna Stephens <rosannastephens@...> wrote:

Hello everyone,

I'm building an enterprise search strategy and am trying to prove or disprove the assumption that younger generations prefer search vs. browse in a consumer and enterprise information search context. I've attempted to research this lightly through our internal resources and am mostly finding content related to marketers or very old research. Is anyone aware of recent research on this topic that you could point me to, or is anyone currently conducting research here?

Thank you!
Rosanna Stephens





Patrick Lambe
 

Hi Rosanna 

I’m sceptical that anything short of a very large scale highly controlled trial would provide any clear answers (and those are blue moon events in our field unfortunately). There are too many confounding factors, including the users’ familiarity with the domain, the relative heterogeneity/homogeneity of the content, the relative heterogeneity/homogeneity of the user communities, the relative stability of the vocabulary associated with the domain (is it a stable technical language or does the language of the domain contain lots of ambiguity, fluidity, synonyms)? All of these factors tend to nudge people towards or away from vocab-based search or structure-based browse.

As Matt points out also, users’ prior experience of the browse/search experience also influences attitudes and preferences. “We want the Google experience” is a common expressed preference but it is usually uninformed by what it takes to create that experience in search, and of how that “under the hood” effort and capability almost never translates into the enterprise. So even an expressed preference is often not actionable, because the context from which the preference comes is not translatable.

By “very large scale highly controlled trial” I doubt anything short of a large enterprise scale trial with hundreds of thousands of people working off different versions of the same environment would actually produce meaningful results. Anything short of that is going to be opinion, based on flimsy anecdotal evidence and usually not accounting for the confounding factors.

We’re much better off with an inductive approach, starting with our specific audience(s), looking at their current environment using references such as those Matt describes, getting opinions, and seeking to make tangible improvements. I personally think (warning: opinion) generational factors count for very little, but nature of the content and its associated language, working context, and satisfaction with current environment count for a lot.

P

Patrick Lambe
Partner
+65 62210383



twitter: @plambesg

Knowledge mapping made easy: www.aithinsoftware.com

On 18 Apr 2020, at 8:09 AM, Matt Moore via groups.io <innotecture@...> wrote:

Hi Rosanna,

Here are my comments (apologies if I am telling you what you already know)
- The consumer internet and enterprise information environments are very different and what is observed in one does not necessarily apply in the other.
- In Australia, every organisation that I have worked in, enterprise search has been poorly implemented (altho to be fair, one place did it OK for a while). There are organisations that do this well but they are the exception.
-  My suspicion is that information foraging behaviours within organisations are less driven by age and more by the quality of the tools available. i.e. whether or not you use search is more dependent on whether the search actually works rather than whether you fit into any generational category.

The best resources on enterprise search are Martin White's 2015 book and the surveys from Findwise (which sadly they have discontinued).

A quick Google yields a fair bit of research on how "young people" use search vs browsing on the internet - e.g. https://www.nngroup.com/articles/young-adults-ux/

Regards,

Matt

On Saturday, 18 April 2020, 09:24:30 am AEST, Rosanna Stephens <rosannastephens@...> wrote:

Hello everyone,

I'm building an enterprise search strategy and am trying to prove or disprove the assumption that younger generations prefer search vs. browse in a consumer and enterprise information search context. I've attempted to research this lightly through our internal resources and am mostly finding content related to marketers or very old research. Is anyone aware of recent research on this topic that you could point me to, or is anyone currently conducting research here?

Thank you!
Rosanna Stephens






Matt Moore <innotecture@...>
 

Hi Susan,

There's a lot going on in that first point!
- Because the public internet and private corporate information environments are so different, you can't simply take techniques that you in one and move them to the other - e.g. Google PageRank works OK in a hyperlinked environment (like the web). The enterprise environment has far fewer links so PageRank doesn't work nearly as well. Google Search Appliance did not use PageRank - it mostly relied on commonly-used relevance ranking algorithms.
- I would say that use cases (what people are trying to do) are also different - so the starting behaviours are also different.

The first two chapters of the book by White (esp. the section "Why can't our search be like Google?") go into this in a lot of detail and are well worth reading - including summaries of the research available.

Regards,

Matt

On Saturday, 18 April 2020, 11:28:05 am AEST, Susan Ostreicher <susan.ostreicher@gmail.com> wrote: 

Hi Matt, 

Thanks for the book and research recommendations - these are very helpful! 

I wondered if you could expand your first point. Do you mean that users prefer to act differently in the consumer internet vs. the enterprise? Or that they need to change their behavior because the two environments are so different? (Or something else?) Do you know of any research backing this up? 

On Fri, Apr 17, 2020 at 8:09 PM Matt Moore via groups.io <innotecture=yahoo.com@groups.io> wrote:
Hi Rosanna,

Here are my comments (apologies if I am telling you what you already know)
- The consumer internet and enterprise information environments are very different and what is observed in one does not necessarily apply in the other.
- In Australia, every organisation that I have worked in, enterprise search has been poorly implemented (altho to be fair, one place did it OK for a while). There are organisations that do this well but they are the exception.
-  My suspicion is that information foraging behaviours within organisations are less driven by age and more by the quality of the tools available. i.e. whether or not you use search is more dependent on whether the search actually works rather than whether you fit into any generational category.

The best resources on enterprise search are Martin White's 2015 book and the surveys from Findwise (which sadly they have discontinued).

A quick Google yields a fair bit of research on how "young people" use search vs browsing on the internet - e.g. https://www.nngroup.com/articles/young-adults-ux/

Regards,

Matt

On Saturday, 18 April 2020, 09:24:30 am AEST, Rosanna Stephens <rosannastephens@gmail.com> wrote:

Hello everyone,

I'm building an enterprise search strategy and am trying to prove or disprove the assumption that younger generations prefer search vs. browse in a consumer and enterprise information search context. I've attempted to research this lightly through our internal resources and am mostly finding content related to marketers or very old research. Is anyone aware of recent research on this topic that you could point me to, or is anyone currently conducting research here?

Thank you!
Rosanna Stephens





David Eddy
 

Matt - 

>
Google PageRank works OK in a hyperlinked environment
>

I deal with “Google behind the firewall.”  

Meaning: how to efficiently research the depths/maze of connections in the decades of operational software applications that are the firm’s (poorly documented) nervous system.

In many software environments the linkages between components is something of a challenge to chase automatically.

I look at the simplicity of explicit HTML links as cheating.  

Inside the firm’s firewall(s) & operational silos, explicit connections are largely not there.

There’s a solid reason for Google to drop their GSA (Google Search Appliance) back in 2016.  It simply does not work on much of the material inside a firm.

- David



Stephen Bounds
 

Hey David,

Microsoft seems to be doubling down on AI an graph discovery as an alternative to traditional search techniques:
https://techcommunity.microsoft.com/t5/microsoft-search-blog/microsoft-search-bringing-intelligence-to-your-world-of/ba-p/960144

I remain skeptical of general purpose AI as a solution for this kind of problem. I feel like organisations that value timely discovery have to invest the time in building the processes and common knowledge necessary to enable it.

To the extent that AI may be a solution, I feel like natural language interfaces (AI chat) have a better long-term shot at success. Bots need context too!

What do you think?

Cheers,
Stephen.

====================================
Stephen Bounds
Executive, Information Management
Cordelta
E: stephen.bounds@...
M: 0401 829 096
====================================
On 19/04/2020 12:23 am, David Eddy wrote:

Matt - 

>
Google PageRank works OK in a hyperlinked environment
>

I deal with “Google behind the firewall.”  

Meaning: how to efficiently research the depths/maze of connections in the decades of operational software applications that are the firm’s (poorly documented) nervous system.

In many software environments the linkages between components is something of a challenge to chase automatically.

I look at the simplicity of explicit HTML links as cheating.  

Inside the firm’s firewall(s) & operational silos, explicit connections are largely not there.

There’s a solid reason for Google to drop their GSA (Google Search Appliance) back in 2016.  It simply does not work on much of the material inside a firm.

- David



Matt Moore <innotecture@...>
 

Stephen,

My 2c is that we don't have a problem with search technology ,we have a problem with information governance - or to put it more bluntly, people giving a ****.

Typically a search engine is only as good as the underlying content. HR policies are the most commonly searched for internal enterprise content - but they actually need to exist for people to find them. On the other hand, if you find 2 undated, mutually contradictory policies, that is also unhelpful. And yet both those scenarios (too few and too many) are fairly common.

Of course, people shoot the messenger - so "search" gets the blame rather than the underlying content and its delinquent owners.

AI can no more solve that problem that it can magically make me lose 10kg.

I do think that some of the AI efforts are interesting.useful and worth watching (and also worrying from a privacy perspective) but this is a case of using technology to distract ourselves from some simple problems. So I think we ultimately agree.

Regards,

Matt

On Sunday, 19 April 2020, 09:43:16 am AEST, Stephen Bounds <km@bounds.net.au> wrote:

Hey David,

Microsoft seems to be doubling down on AI an graph discovery as an alternative to traditional search techniques:
https://techcommunity.microsoft.com/t5/microsoft-search-blog/microsoft-search-bringing-intelligence-to-your-world-of/ba-p/960144

I remain skeptical of general purpose AI as a solution for this kind of problem. I feel like organisations that value timely discovery have to invest the time in building the processes and common knowledge necessary to enable it.

To the extent that AI may be a solution, I feel like natural language interfaces (AI chat) have a better long-term shot at success. Bots need context too!

What do you think?

Cheers,
Stephen.

====================================
Stephen Bounds
Executive, Information Management
Cordelta
E: stephen.bounds@cordelta.com
M: 0401 829 096
====================================
On 19/04/2020 12:23 am, David Eddy wrote:


  
Matt - 



 Google PageRank works OK in a hyperlinked environment


I deal with “Google behind the firewall.”  



Meaning: how to efficiently research the depths/maze of connections in the decades of operational software applications that are the firm’s (poorly documented) nervous system.



In many software environments the linkages between components is something of a challenge to chase automatically.



I look at the simplicity of explicit HTML links as cheating.  



Inside the firm’s firewall(s) & operational silos, explicit connections are largely not there.



There’s a solid reason for Google to drop their GSA (Google Search Appliance) back in 2016.  It simply does not work on much of the material inside a firm.



- David


Tom Barfield
 

Hello Rosanna - I believe my team at Accenture did some research on this (probably 5-10 years ago).  Going off of memory - yes, some people did prefer search or browse.  I do not think we were able to find differences by age - or any other demographic factor.

My team spent quite a bit of time over many years (over 15) working on improving knowledge search quality.  We had quite a bit of positive impact - and it was never good enough. My team was never satisfied, our end users were never satisfied. There are so many angles to consider in improving:  Change the search engine (we did that a few times), Tune the engine (never stop), Too much/too little content, too much/too little time spent on curation. 

I did have a great deal of satisfaction when we made an improvement - and it was great to be able to see how changes we made had impact on our search metrics.  It was just a lot of work.

Tom


Nirmala Palaniappan
 

True. It's ultimately about the quality of content and information. Though, a really poor search engine can make users want to tear their hair out. The common cry is always, "Why can't we have something like Google?".

I think it's critical to complement a good search engine with an intuitive interface, a personalised dashboard and social elements that help push information to users even before they begin to look for it. Browse should, ideally, in my opinion, be made redundant.

Regards
Nirmala 

On Sun, 19 Apr 2020, 05:43 Matt Moore via groups.io, <innotecture=yahoo.com@groups.io> wrote:
Stephen,

My 2c is that we don't have a problem with search technology ,we have a problem with information governance - or to put it more bluntly, people giving a ****.

Typically a search engine is only as good as the underlying content. HR policies are the most commonly searched for internal enterprise content - but they actually need to exist for people to find them. On the other hand, if you find 2 undated, mutually contradictory policies, that is also unhelpful. And yet both those scenarios (too few and too many) are fairly common.

Of course, people shoot the messenger - so "search" gets the blame rather than the underlying content and its delinquent owners.

AI can no more solve that problem that it can magically make me lose 10kg.

I do think that some of the AI efforts are interesting.useful and worth watching (and also worrying from a privacy perspective) but this is a case of using technology to distract ourselves from some simple problems. So I think we ultimately agree.

Regards,

Matt

On Sunday, 19 April 2020, 09:43:16 am AEST, Stephen Bounds <km@...> wrote:

Hey David,

Microsoft seems to be doubling down on AI an graph discovery as an alternative to traditional search techniques:
https://techcommunity.microsoft.com/t5/microsoft-search-blog/microsoft-search-bringing-intelligence-to-your-world-of/ba-p/960144

I remain skeptical of general purpose AI as a solution for this kind of problem. I feel like organisations that value timely discovery have to invest the time in building the processes and common knowledge necessary to enable it.

To the extent that AI may be a solution, I feel like natural language interfaces (AI chat) have a better long-term shot at success. Bots need context too!

What do you think?

Cheers,
Stephen.

====================================
Stephen Bounds
Executive, Information Management
Cordelta
E: stephen.bounds@...
M: 0401 829 096
====================================
On 19/04/2020 12:23 am, David Eddy wrote:


>  
Matt - 



>

> Google PageRank works OK in a hyperlinked environment

>



I deal with “Google behind the firewall.”  



Meaning: how to efficiently research the depths/maze of connections in the decades of operational software applications that are the firm’s (poorly documented) nervous system.



In many software environments the linkages between components is something of a challenge to chase automatically.



I look at the simplicity of explicit HTML links as cheating.  



Inside the firm’s firewall(s) & operational silos, explicit connections are largely not there.



There’s a solid reason for Google to drop their GSA (Google Search Appliance) back in 2016.  It simply does not work on much of the material inside a firm.



- David












Patrick Lambe
 

Greetings Nirmala!

“Browse should be made redundant”

So search is typically language based, and relies on accurately predicting which content is likely to best match which queries.

Browse is structure-based, and structure gives benefits that language does not. If you have a well designed structure (eg. where related items are clustered together irrespective of the language used to describe them, then people who don’t know exactly what they are looking for can start at the landscape view and drill down or find something close to what they want, and look around it to find associated items - much as we do in an unfamiliar supermarket). So browse using a well designed structure can be educational (it can illustrate how things are related) and it can support discovery in ways that query based search can’t. So search and browse should complement each other.

Then there are associations based on usage, in your social example, eg. in the Microsoft graph example. These associations can also support discovery (within the limits of your social graph), but they are typically blind to meaning - i.e. they are not educational - and they are only as strong as the usage patterns already laid down and being analysed.

I agree strongly with the idea of providing for push, and all the mechanisms can in principle support this - search (language you use), browse (context you are in), social (associations you have).

P

Patrick Lambe
Partner
+65 62210383



twitter: @plambesg

Knowledge mapping made easy: www.aithinsoftware.com

On 19 Apr 2020, at 9:38 AM, Nirmala Palaniappan <Nirmala.pal@...> wrote:

True. It's ultimately about the quality of content and information. Though, a really poor search engine can make users want to tear their hair out. The common cry is always, "Why can't we have something like Google?".

I think it's critical to complement a good search engine with an intuitive interface, a personalised dashboard and social elements that help push information to users even before they begin to look for it. Browse should, ideally, in my opinion, be made redundant.

Regards
Nirmala 

On Sun, 19 Apr 2020, 05:43 Matt Moore via groups.io, <innotecture=yahoo.com@groups.io> wrote:
Stephen,

My 2c is that we don't have a problem with search technology ,we have a problem with information governance - or to put it more bluntly, people giving a ****.

Typically a search engine is only as good as the underlying content. HR policies are the most commonly searched for internal enterprise content - but they actually need to exist for people to find them. On the other hand, if you find 2 undated, mutually contradictory policies, that is also unhelpful. And yet both those scenarios (too few and too many) are fairly common.

Of course, people shoot the messenger - so "search" gets the blame rather than the underlying content and its delinquent owners.

AI can no more solve that problem that it can magically make me lose 10kg.

I do think that some of the AI efforts are interesting.useful and worth watching (and also worrying from a privacy perspective) but this is a case of using technology to distract ourselves from some simple problems. So I think we ultimately agree.

Regards,

Matt

On Sunday, 19 April 2020, 09:43:16 am AEST, Stephen Bounds <km@...> wrote: 

Hey David,

Microsoft seems to be doubling down on AI an graph discovery as an alternative to traditional search techniques:
https://techcommunity.microsoft.com/t5/microsoft-search-blog/microsoft-search-bringing-intelligence-to-your-world-of/ba-p/960144

I remain skeptical of general purpose AI as a solution for this kind of problem. I feel like organisations that value timely discovery have to invest the time in building the processes and common knowledge necessary to enable it.

To the extent that AI may be a solution, I feel like natural language interfaces (AI chat) have a better long-term shot at success. Bots need context too!

What do you think? 

Cheers,
Stephen.

====================================
Stephen Bounds
Executive, Information Management
Cordelta
E: stephen.bounds@...
M: 0401 829 096
====================================
On 19/04/2020 12:23 am, David Eddy wrote:


>  
Matt -  



>

> Google PageRank works OK in a hyperlinked environment

>



I deal with “Google behind the firewall.”  



Meaning: how to efficiently research the depths/maze of connections in the decades of operational software applications that are the firm’s (poorly documented) nervous system.



In many software environments the linkages between components is something of a challenge to chase automatically.



I look at the simplicity of explicit HTML links as cheating.  



Inside the firm’s firewall(s) & operational silos, explicit connections are largely not there.



There’s a solid reason for Google to drop their GSA (Google Search Appliance) back in 2016.  It simply does not work on much of the material inside a firm.



- David













Nirmala Palaniappan
 

Hi Patrick, 

:-) You are right. An unfamiliar supermarket is a great example to explain why Browse may be essential for serendipitous discovery.
I can't help but hope that personalized dashboards backed by intelligent software and algorithms based on collective behaviour will somehow bring up everything that a person may need in the long run. Browse is perfect for a situation that involves reflective and exploratory learning. (I am influenced by corporate cultures that seem to rarely find the time to explore and are largely in reactive modes) 

Regards

On Sun, 19 Apr 2020 at 07:30, Patrick Lambe <plambe@...> wrote:
Greetings Nirmala!

“Browse should be made redundant”

So search is typically language based, and relies on accurately predicting which content is likely to best match which queries.

Browse is structure-based, and structure gives benefits that language does not. If you have a well designed structure (eg. where related items are clustered together irrespective of the language used to describe them, then people who don’t know exactly what they are looking for can start at the landscape view and drill down or find something close to what they want, and look around it to find associated items - much as we do in an unfamiliar supermarket). So browse using a well designed structure can be educational (it can illustrate how things are related) and it can support discovery in ways that query based search can’t. So search and browse should complement each other.

Then there are associations based on usage, in your social example, eg. in the Microsoft graph example. These associations can also support discovery (within the limits of your social graph), but they are typically blind to meaning - i.e. they are not educational - and they are only as strong as the usage patterns already laid down and being analysed.

I agree strongly with the idea of providing for push, and all the mechanisms can in principle support this - search (language you use), browse (context you are in), social (associations you have).

P

Patrick Lambe
Partner
+65 62210383



twitter: @plambesg

Knowledge mapping made easy: www.aithinsoftware.com

On 19 Apr 2020, at 9:38 AM, Nirmala Palaniappan <Nirmala.pal@...> wrote:

True. It's ultimately about the quality of content and information. Though, a really poor search engine can make users want to tear their hair out. The common cry is always, "Why can't we have something like Google?".

I think it's critical to complement a good search engine with an intuitive interface, a personalised dashboard and social elements that help push information to users even before they begin to look for it. Browse should, ideally, in my opinion, be made redundant.

Regards
Nirmala 

On Sun, 19 Apr 2020, 05:43 Matt Moore via groups.io, <innotecture=yahoo.com@groups.io> wrote:
Stephen,

My 2c is that we don't have a problem with search technology ,we have a problem with information governance - or to put it more bluntly, people giving a ****.

Typically a search engine is only as good as the underlying content. HR policies are the most commonly searched for internal enterprise content - but they actually need to exist for people to find them. On the other hand, if you find 2 undated, mutually contradictory policies, that is also unhelpful. And yet both those scenarios (too few and too many) are fairly common.

Of course, people shoot the messenger - so "search" gets the blame rather than the underlying content and its delinquent owners.

AI can no more solve that problem that it can magically make me lose 10kg.

I do think that some of the AI efforts are interesting.useful and worth watching (and also worrying from a privacy perspective) but this is a case of using technology to distract ourselves from some simple problems. So I think we ultimately agree.

Regards,

Matt

On Sunday, 19 April 2020, 09:43:16 am AEST, Stephen Bounds <km@...> wrote: 

Hey David,

Microsoft seems to be doubling down on AI an graph discovery as an alternative to traditional search techniques:
https://techcommunity.microsoft.com/t5/microsoft-search-blog/microsoft-search-bringing-intelligence-to-your-world-of/ba-p/960144

I remain skeptical of general purpose AI as a solution for this kind of problem. I feel like organisations that value timely discovery have to invest the time in building the processes and common knowledge necessary to enable it.

To the extent that AI may be a solution, I feel like natural language interfaces (AI chat) have a better long-term shot at success. Bots need context too!

What do you think? 

Cheers,
Stephen.

====================================
Stephen Bounds
Executive, Information Management
Cordelta
E: stephen.bounds@...
M: 0401 829 096
====================================
On 19/04/2020 12:23 am, David Eddy wrote:


>  
Matt -  



>

> Google PageRank works OK in a hyperlinked environment

>



I deal with “Google behind the firewall.”  



Meaning: how to efficiently research the depths/maze of connections in the decades of operational software applications that are the firm’s (poorly documented) nervous system.



In many software environments the linkages between components is something of a challenge to chase automatically.



I look at the simplicity of explicit HTML links as cheating.  



Inside the firm’s firewall(s) & operational silos, explicit connections are largely not there.



There’s a solid reason for Google to drop their GSA (Google Search Appliance) back in 2016.  It simply does not work on much of the material inside a firm.



- David














--
"The faithful see the invisible, believe the incredible and then receive the impossible" - Anonymous


 

Hi everyone,

I'm just getting back to this thread now and reading carefully through your responses and recommendations. Thank you so much for sharing your knowledge and taking the time to respond!

Matt, I really appreciate the tip about Findwise. I looked up their last survey and found some great ideas in there that will help steer the direction I'll be taking with our strategy. I'll have to look up Martin White and take some time to read his book. I completely agree with you regarding information governance, and I'm fairly certain at this point that a large portion of my strategy will be recommendations about information governance. As I work through focus groups and interviews at my company, it's very clear that we don't capture critical knowledge, and this is why search is unable to find it. If we do capture it, it's hard to find AND use because the content lacks any sort of standard and is often outdated.

Patrick, I completely agree that targeted research is a much more useful measure, and we're working on this using exploratory research with our employees through focus groups, interviews, and later surveys and data analysis. I was curious where this generational assumption came from, because people have been talking about it like a known fact, and it sounds like it was just a conjecture by someone uninformed or observing behavior of one person they knew (or even personal preference, as seems to come up often in my team).

Stephen, yes, chatbots are useless without context or even without knowledge from which to pull. We have so many teams jumping to chatbots, especially in this COVID-19 work from home situation, and they don't release they haven't captured anything for the chatbot to use.

Tom, do you happen to have your research in paper (or electronic paper) format that you could share?

Again, much gratitude to this group. This discussion will help me tremendously in the coming weeks.

Take care, and have a great weekend!

Rosanna Stephens