from my point of view  its not easy to use demographic indicators to measure the process of development, you need an economic measure such as GDP growth, or GDP per capita growth, and i think its very easy to find information about that since 1950s. Its not easy to avoid economic measure to measuring economic development using Only demographic factors you have to join Human Indicators with Economic Indicators .......  

per capita GDP is a good indicator you can download the time series data using World Bank development Indicator (WDI).

I was thinking of historical times, when there is no statistical data (eg. the Middle Ages) and the only data that can be found is for example: number of people.

There is a source for measuring economic growth for historical data. It has been built by Maddison and it is free to download here http://www.ggdc.net/maddison/maddison-project/data.htm . It is widely used for investigation in historical perspective, as your case is.

As an alternative, especially if you are not focused on countries, you can indeed use demographic data as a proxy for economic development. The larger the population, the higher the production. In particular, you may want to use demographic indicators for towns and cities(which, if I remember correctly, is one of the indicators accounted for by Maddison himself), since they were (and are) the main centers for economic activities such as commerce, proto industrial production such as artisanal manufacturing.

Then, if you mean development in a broad sense, then using historical data related to economic growth should solve your problem, since I would feel safe to overlap growth and development concepts until the first industrial revolution.

Roberto, thank you very much for your comment.

(1) A conceptual note: since you are tackling non-market societies, economic development in times such as the Middle Ages or ancient slave societies is better understood as expansion and/or decrease of output or productivity in certain key-areas (esp.  agriculture) rather than the much broader economic growth that we use for today's economies.

(2) If you are interested in the Middle Ages, I strongly recommend you to make use of the research by Fernand Braudel and the team of the Annales school. Methodologically, they provide as good as an example you will ever find of thorough empirical study of historical societies and their economies; theoretically, they avoid the cliometricians' basic error of approaching previous economies as though they were contemporary market economies. 

My comment may appear rhetorical, but it isn't. I might say that the first question that we need to ask is: what is economic development?

Baruchello and Martino positions can be read as akin to the question I am doing.

As long as you have defined what do you mean by economic development, it will be easier for you to answer your main question.

Thank you for the suggestions. I'm interested in the indicator available for long stretches of time. Indicator which does not change through time and space. In the case of France are easily accessible data. I used data on the number of city dwellers https://www.researchgate.net/publication/224968384_Zmiany_struktury_osadniczej_Francji. However, in the case of Eastern European countries with statistical data is a big problem (no censuses, frequent change of boundaries and loss of data during the wars). Therefore, my question is about a universal and accessible the indicator development.

Article Zmiany struktury osadniczej Francji

The source you cite is about recent times, for which we do have quite a wealth of information. As for older ones, there is no straightforward, universal and accessible indicator of economic development: one has to start with the awareness of how incomplete the information is and then delve into the available sources, as fragmentary as they are, yet avoiding sweeping generalisations based on cherry-picked data that fit one's favourite model. That is the job of honest economic historians, such as Braudel and his peers.  

http://www.tandfonline.com/doi/abs/10.1080/10848770.2013.804705

Giorgio, thank you very much for your comments.

I think we should consider such analysis in a "stage-theory" fashion. We will use the economic data to describe the important characteristic in the stages. Note that according to Lionel Robbins, such an approach has little predictive power. (See Robbins, "The Theory of Economic Development in the History of Economic Thought, "Macmillan, 1968,p.3)

Methods of measuring economic development in historic times in absence of  detailed statistical data would be country specific. Also the cut off point to consider as historical times too is to be spelled out and decided first..  The methodology needs to be developed on the basis of  data available and not available for the social and economic indicators of development.   For indicators where data in published form are not available, it has to be collected from the old records or files if preserved.  In fact stories and folklore also reflects the socioeconomic conditions of historic times. .....  .....