I think it depends on many factors such as the education, the wealth, age, availability of services on the Internet in a given country or area

You can explore correlations between different data. Were such research?

Studies (for example by the Pew Research Center) have shown across higher income nations that it is not money which sets the limit, it is user (that is, non-user) preferences and perceptions of need.  For example some rural, elderly Japanese see no need or benefit from Internet access; and of course they must be right.

So the present 'saturation point' varies from around 85% to a max 95% in advanced nations; which at the higher end is the same as the highest level fixed line telephony penetration reached (on a household basis) back in the day.

However, if we changed the question to - what is the saturation limit for mobile connectivity; which includes already in advanced nations and many developing nations as a standard feature mobile internet access - there the saturation point is OVER 100%, since many of those same elderly Japanese rural residents would be in a panic if they did not have a cell phone in their pocket and were unable to reach their family or emergency services.

So, present forecast is - 4.5bn mobiles worldwide presently, which is roughly 64% of the world's population, trending towards 70% by 2017.  By 2017 50% of those will be smartphones with - wireless broadband Internet access capability.  NOT saturation by any measure, even if income and other limits have slowed the rate of growth.

Therefore, combining fixed and mobile broadband, and noting that some who cannot have their own devices in low-income and rural areas may occasionally - rent- access whether via a formal Grameenphone-type  program, or through more informal entrepreneurial efforts of community residents - it is entirely reasonable to expect that by 2025, your question would look amusingly outdated in thinking: there is NO Internet saturation limit, only limits to people's income, education, and infrastructure accessibility.

Especially because - you forgot to ask about the Things around us, and how many of them - want and desire? ; ) or need? : ) Internet access. Or rather, their owners/operators/users wish to interconnect via the Internet.

Cisco estimated 25 billion things are connected to the Internet - this year; rising to 50 billion by 2020.

That is - already today, per capita as a global average, there are more than 3 times as many - Things - connected to the Internet as people on earth. Or if we prefer, we can say that per - smartphone - user, there are about 10 Things connected to the net today.  Or, for the estimated 3.1 billion Internet users today, each of them has - 8 things connected to the net. Today.

To conclude then, there is NO saturation limit, for people or Things on the Internet.

There are preferences and individual choices, and limits to income, education, access, and infrastructure. Today, and tomorrow; varying also by geographic, cultural,and other variables.  Which may ultimately lead to that same magic 'saturation' number of 95% recurring. For people, but not for Things.

Until...human-embedded health sensors require all of us to be - always on ; )

In say, 2050.

Hi Lee

Thank You for interesting information.

A report by the UN's International Telecommunications Union (ITU) said 44 %  of the world's households will have internet access by end of 2014. Close to 78 % in developed countries will be connected to the Internet, compared to 31 %of households in developing countries. The figures show that household Internet access is approaching saturation levels in developed countries. Reaching a 90% or more does not require a long time since many services have given up or will give up, soon, the paper versions of their notifications or bills. This way, more people will find it difficult to live without Internet. As for e-learning & on-line exams, more educational institutes will be encouraging these methods because they are cost-reducing so all university student will be obliged to connect with the Internet.  

Hi Nizar

Thank You for answer. The saturation level is important  to define the level of the digital divide. It seems to me that impact of age will be decreasing. I think that 90% of saturation is not possible. But about 80%  is possible.

In 1966, there were less than 10 black & white TVs in my town (which counted then ~ 40,000) but ,at present, every house has, at least,one colorful TV (with a population that reached  ~ 80,000). Besides PCs, many persons, in our 3rd world country, are connected to the internet by mobile phones, ipads,tablets...etc. I may subtract the small children & the disabled elderly from the internet invasion(!) but these are less than 10% in developed "ageing" countries. The percentage, in my opinion, remains to be seen in few years time & it may be a number between your estimation & mine ! :).

Hi Nizar

Have You got any timeseries with data?

 Dear Dr. Jan Grzegorek,

Sorry that I have not seen your question before this moment. A scientific timeseries with data is difficult to gather in our country because of its unique conditions. Only estimation, based upon the observations of the living elderly persons, can work in our case. I really hope that a young researcher can go into this area of research & collect through interviews data from which good conclusions are drawn. My regards.